Episode 1111 1111 — October 15th, 2026 — Available at read.fluxcollective.org/p/255

Contributors to this issue: Erika Rice Scherpelz, Justin Quimby, Neel Mehta, Boris Smus, MK

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Laura Eberly, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“You know how in video games, if you get the super-duper high score, it eventually flips back to zero? Well, sometimes Homestar does something so stupid, he flips back to smart.”

— Strong Bad, “4 branches,” Strong Bad Email #161 (2006)

📝 Editor’s note: This is episode 255. In honor of the maximum value that can be stored in a single unsigned byte, we’ve decided to theme this episode around the concept of “fullness.” We hope you enjoy it.

💯😵‍💫 Dying at capacity

We’ve noticed waste at our grocery store: open registers with no line, food prepped at the deli that doesn’t get sold, a steady rotation of an employee or two who don’t seem to have anything to do. Clearly, we need to up efficiency. So we cut the number of employees in the store by 25%. Deli items are held to a strict “no overage” policy. And if a register has no line for more than five minutes, we send the employee elsewhere. A couple weeks later, register lines are long enough that they start running into each other. Customers complain of shortages at the deli. Deliveries start piling up because they can’t be shelved quickly enough.

This probably seems obvious. When you take all the slack out of a system, it becomes impossible to respond to natural variance in the system. That deli overage (which, it turns out, was only a small percentage of the day’s throughput) is the price to pay for having product available on demand. The idle cashiers and slight excess of staffing allow you to deal with the variable rates at which products flow in and out of the store.

What is more surprising is how much slack a system needs to operate well. It turns out that a system can start degrading at somewhere between 70% and 90% capacity. This is the realm of queueing theory. To give a brief overview: In any system where work flows in and has a time delay before completion, work will start to build up whenever the inflow exceeds the rate at which new work can be absorbed into a system. If work arrives steadily and is processed steadily, we can understand the behavior of our queues exactly. But in most real queues, both the input rate and the processing rate vary. For our grocery store example, customers may arrive in bursts (e.g., shopping after work) and cashiers have different speeds at which they can scan and bag. (The second is why the self-serve line can often feel frustratingly slow; the amateur cashiers vary greatly in how quickly they can get things from cart to bag.)

As Donald G. Reinertsen derives in The Principles of Product Development Flow, a key variable for a queue is the proportion of time spent waiting for an average item. An item that can proceed straight to work spends 0% of its time waiting. An item that waits twice as long as it spends in process spends 66% of its time waiting. Depending on exactly how our queue works, we start seeing dramatic upticks in the proportion of time spent waiting. And as many consumers know from experience, once you start spending more time waiting than on the thing, it’s almost inevitably frustrating.

A more human-centric take on a similar idea comes in Tom DeMarco’s Slack. Although the book’s discussion of Dilbert feels a bit dated, DeMarco’s basic thesis is still sound: companies try to maximize efficiency, but by doing so they remove their ability to respond nimbly to changing conditions. Maximizing efficiency feels like it will allow a company to move most quickly, but both variability and unpredictability mean that too much efficiency can trap us in a corner. We all need more slack in our lives (and yes, we consider it ironic that the Slack app reduces slack).

Queues and variability are a fact of life; after all, we don’t want people sitting around bored or idle. So what can we do about it? Reinertsen and DeMarco both give us some clues: first, don’t try to look at utilization. It can be hard to measure well and doesn’t provide much actionable guidance. Instead, monitor queue length and wait time. These can provide useful guidance about capacity. Additionally, we can also be thoughtful about how we take items from the queue. First in, first out (FIFO) can work if queue items have comparable economic value and expected processing time. However, once we introduce variability, it can help to take that into account. For example, an emergency room triage system ensures that the most “valuable” cases — the critical injuries — are processed most quickly. On the processing side, processing a number of quick-delivery items can clear out the queue enough to reduce average wait time.

On the human side, DeMarco recommends that we give people capacity to sense and respond to change. Some of this is about culture: we must learn to embrace risk and exploration. It’s okay if not every piece of work generates success. Some advice is about the day-to-day: reduce low-value busywork that keeps people occupied but prevents them from having time to explore. Yet at the base of it all is attitude: humans are not machines. We should respect and value the need for downtime, not as waste but as a reset.

So as we, on episode 255, fill out our first 8 bits of capacity, we want to take a moment to respect the spare capacity that allows us to keep producing this newsletter and allows you to spend some of your precious time reading it.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🍧 Hackers heated up fridges at French supermarkets, spoiling food and causing fires

150 French grocery stores were hit by hackers who raised the temperatures of their refrigerators and freezers to “as high as 50 degrees Celsius (122 degrees Fahrenheit),” which spoiled food, started fires, and forced refrigerated food aisles to close. Many of these fridges were connected to the internet to let maintenance companies remotely adjust temperatures, but they were running “outdated firmware” that left them vulnerable. The attacks reportedly resemble recent fridge hacks at grocery stores on US military bases; these “commissaries” are a key benefit for military families and veterans, giving them substantial discounts over civilian stores and a boost to morale.

🚏🇮🇩 Indonesians are being recruited into the Russian army over social media

Social media recruiters are messaging Indonesians and offering them jobs as cooks, drivers, cleaners, nurses, and more in Russia, even offering to pay for airfare and e-visas. But, according to their families, dozens of young people were instead pressed into signing contracts (in the Russian language) to join the Russian army or build drones on the front lines; several couldn’t leave because their ‘sponsors’ took their passports. Indonesia isn’t the only country with this problem: India, Sri Lanka, Nepal, Colombia, Vietnam, and several African nations (among many others in the Global South) have had problems with their citizens getting duped into joining the Russian army.

🚏📲 Norway plans to temporarily ban smart glasses in public

Norway is planning a temporary ban on smart glasses with audio/video recording functionality in public places such as “parks, beaches, museums, shopping centers,” and more, citing concerns about people being recorded without their knowledge. Over in the UK, some cinemas and pub chains are banning smart glasses on their premises, while the Australian government is weighing a ban on smart glasses in government offices.

🚏🪃 A YouTube political commentator used AI to fake an Aussie accent and talk politics

A YouTube channel called Oceania Unfiltered racked up over 3 million video views talking about Australian and New Zealand politics. Its host appeared to talk with a thick Aussie accent, but it turned out he’s a Brazilian man who used AI to change his voice and edit the video to make his lip movements match. He used his platform to promote a far-right Australian party and got thousands of comments from real Aussies. Meanwhile, Facebook is also overrun with AI-generated posts about Australian politics, which are largely created by people from Indonesia, Sri Lanka, Vietnam, Bangladesh, etc. When users engage with these posts, the creators get paid in Australian dollars — which convert to a lot of money in these developing countries’ local currencies.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Why AGI Will Not Happen (Tim Dettmers) — Argues that intelligence hits physical walls: human brains are already maxed out (pregnant women can’t metabolically support larger ones), GPUs plateaued in 2018, and linear capability gains now demand exponential resources. As a result, China’s model of “useful AI at reasonable cost” will beat the USA’s unrealistic ambitions of superintelligence, which the author dismisses as an Oxford philosophers’ fantasy. The notion of an AGI that improves itself (thus causing a runaway effect and the ‘singularity’) is a “deeply flawed idea” that “treats intelligence as purely abstract and not grounded in physical reality… to improve any system, you need resources,” and resources get limited quickly.

• Artificial Deadlines: Evidence of Fraud in an Influential Study About Procrastination (Data Colada) — A team of behavioral scientists reveal irregularities in the data from an influential Dan Ariely psychology study. In this post and its follow-up, they use forensic methods to find evidence of data duplication, implausibly large effect sizes, and intentional data alteration to support a hypothesis. One interesting heuristic: participants in the study self-reported the number of minutes they spent on certain tasks, and while a replication showed that people tend to round time intervals to round numbers like multiples of 10 or 15, the original data had a smooth distribution with no round-number bias — suggesting that an algorithm or RNG was used to generate these “human-reported” stats.

• Brand New India (Phenomenal World) — Historian Ravinder Kaur argues that, in developing countries like India, economic liberalization and globalization have counterintuitively gone hand-in-hand with democratic backsliding and ethnonationalism. One mechanism is the commodification of the nation-state, where a country becomes a brand for foreign investment and tourism and thus tends to promote the majority culture in its advertising and publicity campaigns. Also, liberalizing economic reforms tend to require centralization of power as the national government breaks down regional barriers to trade and imposes a single market.

• Is Kinetic Energy Actually ½mv²? (Stand-Up Maths / YouTube) — Mathematician Matt Parker does a satisfying calculus derivation to prove that our familiar equation for kinetic energy, as with most equations from Newtonian physics, is just a first-order approximation for the real relativistic equation, but close enough at low speeds. He also chats with a CERN scientist about how they compute energy amounts at the Large Hadron Collider’s mind-melting speeds.

🔮📬 Postcard from the Future: Inbox Zero Never

A ‘what if’ piece of speculative fiction about a possible future that could result from the systemic forces changing our world.

// What might the impact of the broad adoption of agents have on the general populace’s inboxes?

2028: The interior of a local coffee shop in Westwood, Massachusetts (a town a half hour south of Boston and well outside of the MIT/Harvard University Zone).

Two old friends sit down for a cup of Dunkin on Saturday morning. One looks bedraggled.

“It’s been a minute, Riley! You look awful. What’s going on?”

“It’s my damn email—” <ding - their phone notification rings> “Ever since every store and every Tom, Dick, and Harry in my contact list started using AI, their damn Agents are flooding <ding> my inbox <ding> with messages!”

Casey ruefully shakes their head in acknowledgment. “My younger cousin was telling me why when I was complaining about the same thing. Email is apparently this ‘universal protocol’ that all the AIs can use to talk to each other. So now our inboxes are turning into their switchboards.”

“Yeah! I thought the old spam and political emails were bad, now I’m sorting through dozens of messages a day! I used to only get ten emails a day… those were the good old days!” <ding> “And now Google, Microsoft, and Yahoo are charging more money for their ‘custom personal Agents’ to sort the whole thing out. I’m not paying them more for a problem they created!”

They both sigh and take a sip of their coffee.

<ding>

“Riley, at least turn off those damn notifications!”

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Out of the frying pan, into the… water? NYC on the Hudson River at sunset. // Photo: Neel Mehta, FLUX

Episode 254 — October 8th, 2026 — Available at read.fluxcollective.org/p/254

Contributors to this issue: Justin Quimby, Erika Rice Scherpelz, Ben Mathes, Neel Mehta, MK, Boris Smus

Additional insights from: Ade Oshineye, Dart Lindsley, Jasen Robillard, Jon Lebensold, Laura Eberly, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“The foundation of cooperation is not really trust, but the durability of the relationship.”

— Robert Axelrod, The Evolution of Cooperation (1984)

💰🎲 The nature of the bag

We live in interesting times. As anxiety grows over bond rates, supply-chain disruption, or new plagues, and as government stability ping-pongs through the information ecosystem, we see more and more people become aware of the K-shaped economy. This is an economy where a booming and prosperous middle class becomes a history lesson, not a lived reality. This is a zeitgeist where the phrase “permanent underclass” has blossomed in the past 18 months.

The K-shaped economy changes the nature of day-to-day lived experience, but the impact goes beyond that. In game theory, huge differences exist between players playing an infinite game vs. a finite one. In an iterated infinite game, there’s an incentive to cooperate, at least to some degree. Malicious and destructive tactics accumulate distrust and ill-will from other players, resulting in fewer benefits for the malicious player. In finite games, non-cooperative behavior can result in a player winning significantly more before walking away. Those are short-term gains… but when the short-term is all there is, those are the only gains to be had.

The short-term players “got their bag.” When “the bag” means generational wealth, it can make otherwise distasteful tactics feel justified. The end, one might say, justified the means. But in these short-term gains, “the bag” often comes at a cost to others, exacerbating the differences between the haves and have-nots.

We see this in the crypto and financial worlds. In every crypto rug-pull, a few small groups of insiders are “bag takers,” while the vast majority are left “holding the bag.”

Let’s add a third dynamic: the opportunistic outsider. These are the people who see opportunity in an environment with massive economic upheaval. California’s Gold Rush (1848-1858) saw vast fortunes minted by those who “sold shovels” to prospectors. The aftermath of the US Civil War saw some Northerners, pejoratively called “carpetbaggers” by Southern Democrats, moving South to make their fortunes during the Reconstruction era (1865-1877). One specific example is that, in 1870, Northerners controlled 21% of the South’s railroads (by mileage); 19% of the directors were from the North. By 1890, they controlled 88% of the mileage; 47% of the directors were from the North.1

Unlike those who “got their bag”, the opportunists are not necessarily using immoral means, but they’re often resented nonetheless. They are the outsiders who figured out how to play the game better than those who “earned” their way there.

But not all shovel-sellers are the same. One might make a fortune from the rush and disappear. Or they might make tools that remain useful when the rush is over. They might become part of the community. Both spot an opportunity; what they leave behind is different.

That suggests another dynamic: the builder. Builders may be opportunists, but their opportunity is win-win, not zero (or negative) sum. A general store might open to sell supplies to prospectors, then become the place where people trade news and find work while they buy what they need long after the rush ends. These are choices about the rules of the game. If more people benefit from another round, more people have a reason to keep playing together.

Every gold rush has those who “get the bag” and those who “hold the bag.” It has the opportunists and the builders. This is a dynamic that seems to arise when fortunes are to be made and there are few options between being a winner and a loser. This sort of environment primes people to take risks, the sort of risks that lead to a win or a loss. Many are treating AI as the current gold rush. So as the AI engine continues to churn away and more people chase their bag, consider whether you want to play the finite game or choose a different set of rules and take the odds on a cooperative, infinite game.

Although, perhaps the choice isn’t entirely up to any one player. Be it finite or infinite, a game is something we play together. So in the game of our economy, relationships, and value systems, how might we build tools and institutions that give people a reason and a way to stay in the game for the long run?

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🎨 Someone vibe-coded open-source versions of Adobe’s top seven creative apps

A veteran software developer used Claude Code to develop free, open-source versions of seven Adobe Creative Suite apps: Photoshop, Illustrator, Premiere, Lightroom, Acrobat Pro, After Effects, and InDesign. These apps, built entirely in the popular language Rust, are “clean-room reimplementations” of the famous (and famously expensive) Adobe products, and are compatible with Adobe’s official formats. One reviewer found these apps glitchy but useful and close enough to the originals.

🚏🧲 Germany’s far-right and far-left party teamed up to propose ending Ukraine aid

The recent parliamentary election in the eastern German state of Saxony-Anhalt led to a shakeup: the infamous far-right AfD party won the most seats and fell just a few short of an outright majority. The other mainstream parties show little interest in working with AfD, so it seems to be allying with the far left-wing party BSW; the parties would combine for a slim majority of seats. One of AfD and BSW’s first acts will be proposing a joint resolution that would call for ending military aid to Ukraine and lifting sanctions on Russia. (Observers note that states have no impact on foreign policy, so the resolution would be purely symbolic.) While AfD and BSW may seem to be strange bedfellows, both are Eurosceptic populist parties and are considered to be culturally conservative — which brings horseshoe theory to mind.

🚏🫎 A Colorado wildlife overpass (the world’s largest) reduced animal collisions 91%

Highways can be highly harmful to wild animals, acting as walls that lead to habitat fragmentation; animals that try to cross the highways often collide with cars, endangering beast and human alike. In 2019, Douglas County, Colorado (not far from Denver) built the world’s largest highway overpass: a massive, grass-covered bridge that lets animals walk above the road. The 209-by-200 foot bridge has been credited for reducing wildlife collisions by 91%; the collision reduction means the overpass will pay for itself in 15 years, according to the state’s department of transportation. The overpass was designed for the pronghorn and elk that roam the area, but biologists have noticed that bears, deer, coyote, and other animals have used it as well.

🚏🪙 California is banning public officials from making memecoins

Gavin Newsom, the governor of California, recently signed a bill that will ban the state’s public officials, who includes elected officials and state government employees, from making memecoins, or joke cryptocurrencies whose value depends solely on hype and which are usually intended for quick pump-and-dumps. The bill also bans companies from launching memecoins using “the likeness or image of a public official.” Besides the potential harm to consumers (nearly a million “investors” lost $3.8 billion on a Trump-themed memecoin), the governor warned that memecoins could let public employees corruptly profit off their office.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• How to Write with LLMs (Final Ward) — Proposes two simple rules to make LLMs enhance your writing process rather than “pasteurizing and jacking it with corn syrup”: never use a single word the LLM suggests to you, and avoid encouragement from AI (which leans toward sycophancy). Instead, use LLMs as copyeditors; if they flag a problematic passage, you can ask another fresh LLM to compare the original and revised versions and pick the better one.

• A Series of Vignettes From My Childhood and Early Career (Jason Scheirer) — Argues that every generation confidently predicted programming’s imminent death. In 1996, the author’s mentor warned that OOP would create “a library of objects, like books on a bookshelf” where “business people just snap the object libraries together like LEGOs,” but the work just migrates up one abstraction layer each time.

• Barbarity and Civilization: Explaining Wanton Violence in the Conduct of War (Dead Carl and You) — Examines the US’s current “fruitless bombing campaign against Iran” through the lens of Carl von Clausewitz’s military theories, seeking to explain why a presumably rational modern state would waste its energy and matériel on unproductive violence. Military historians in the Enlightenment thought that advancing “civilization” would lead war to become more refined and less violent, but Clausewitz posits that advancing development just increases states’ potential for violence. Plus, bureaucratic institutions can still commit inefficient violence if their leaders get emotional or if policymakers believes that said violence is actually useful.

• Heart Rate Variability Biofeedback: How and Why Does It Work (Frontiers) — Introduces HRV, an up-and-coming idea in physiology. Turns out the pressure wave produced by the heart and breathing are deeply intertwingled. By breathing at a certain frequency (~6 breaths a minute), one can increase HRV temporarily, and also exercise the neuroplasticity in the baroreflex, increasing resting HRV.

🔍📆 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: ground truthing.

Customers love our new feature! 1,000 uses in the first week. Hurrah! Then someone notices that 900 came from the same hundred accounts. They were all trying repeatedly to finish one task. It turns out the feature wasn’t really good. It was bad. Really bad. So bad it hardly worked.

We frequently turn to the idea of ground truth, not as a noun, but as a verb. This is the moment where a set of beliefs experiences a temporary “mark to market.” Delusions collapse back into empirical data. This collision with ground truth can be organizationally and culturally disruptive since it pierces the veil of load-bearing polite fictions. “We’re doing well.” “I know what’s going on.” “We will survive the next year.”

Sometimes we accidentally (or intentionally) manufacture the need for ground truthing. When we greenshift status so that the news gets just a little bit better each level it goes up the hierarchy, then we are creating the need for ground truthing down the line.

Why only a temporary correction? Because in a world of geologically active ground, some truth is shifting. Many things are stable — this isn’t relativism! — but almost all of us have key parts of our worldview that belong to the more molten aspects of reality. So we ground truth today just to find ourselves wrong tomorrow. For example, it turns out that “Ha ha! AI hallucinates” never became truly false, but agents, LLMs running tools in loops, made hallucinations much, much less relevant. (In fact, you could say that tool use allowed agents to ground truth, thus making this example relevant on two levels.)

To avoid the sort of major corrections that disrupt an organizational culture or a personal belief system, we need to be in a consistent process of building new ground truth. When we do so, then the ground truth can be more easily known, adjusted for, and by its own absorption, become irrelevant. This process cannot be saved up for later. We always have to be building new ground truthing mechanisms, or else our existing mechanisms will calcify.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 253 — October 1st, 2026 — Available at read.fluxcollective.org/p/253

Contributors to this issue: Justin Quimby, Erika Rice Scherpelz, MK, Neel Mehta, Boris Smus

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Laura Eberly, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“That’s too coincidental to be a coincidence.”

— Yogi Berra

🌊⛵ The uses of adversity

After a beautiful summer sailing on a calm lake, you take your wooden boat into rougher seas. Before you know it, the boat you thought was sound is taking on water. Not much. You’re able to get to a dock and patch things up. But still, no captain or crew wants to be in this situation.

Above the usual waterline, sun and wind can dry a wooden boat’s planks. As the wood shrinks, seams open. Everything seems fine… until the boat experiences waves that reach above “normal”. These are fair-weather seams, a niche term but a real phenomenon for those who sail and maintain wooden ships.

Teams also have fair-weather seams. Imagine a group that has shipped reliably for months. Everyone knows what to do and they do it well. Then there’s a weird data corruption bug on a Friday night. The eng team thinks support owns the response. Support thinks eng owns it. The lead assumes an engineer is already looking at it. During the fair weather, the team’s ability to handle the incident atrophied. The good times didn’t just hide the gap. They created it. With no incidents to respond to and nothing to keep the team in practice, they were unprepared when something went wrong.

This is closely related to Taleb’s idea of antifragility. Some systems gain from challenge rather than merely survive it. Muscles get stronger with use. Fair-weather seams illustrate the flip side of antifragility: when a system needs challenges to get better, then seemingly good times have a hidden cost. The good times let weaknesses develop unseen. And we’re never “done” with the challenge. Being exposed to water can help the planks swell and close the seam, but a long enough spell of dry weather and calm waters can let them open again.

In those seemingly fortunate times when all is smooth sailing, it can be worth introducing some friction into our lives. This is the idea behind drills. It is also the idea behind chaos engineering, which involves running failure experiments on a system. This can range from the random failures of a chaos monkey to more structured exercises like Google’s DiRT scenarios.

All this raises a somewhat uncomfortable question about smooth sailing, about the good times. While no one wants to be in constant chaos, the flip side has its own challenges. When nothing is going wrong, how do we know that our seams are tight? We need to take a first step toward discomfort. We have to open ourselves to risk to see how ready we are. As Shakespeare says, “Sweet are the uses of adversity.”

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🇸🇬 Singapore’s government is building a dating app

Singapore’s Government Technology Agency is launching FirstDate, a dating app for “singles aged 21–35.” (It’s currently in beta for government employees only.) It uses similar algorithms to mainstream dating apps, though it relies more on questionnaires than swiping, but the most interesting part is that it verifies accounts through the city-state’s digital ID system, which is supposed to help thwart fake profiles. The launch comes amid a recent finding that Singapore’s birth rate fell to just 0.87, down from an already low 1.24 ten years ago; earlier this year the government tasked a working group with “addressing falling birth rates.” Perhaps this app hopes to chip in.

🚏🍟 McDonald’s is using AI for dynamic pricing

Shoppers have long feared the potential of digital price tags that can adjust product prices on the fly, and it seems like McDonald’s is making them real. The Golden Arches now offers an AI system to franchisees that analyzes customer spending at nearby restaurants and gives advice about what prices to set; it sends franchisees messages like “your restaurant is showing medium sensitivity to price [based on] customer willingness to pay in your area.” McDonald’s maintains that this system is purely advisory and franchisees are free to set their own prices, but McDonald’s corporate has also been known to push prices on franchisees and punish those who deviate too much.

🚏🏎️ Tech workers got ChatGPT to drive a Toyota

LLM benchmarks — bundles of tests that give models scores for performance on certain tasks – are all the rage, and now that LLMs are easily cracking knowledge-based exams, researchers have started releasing benchmarks for things like robotics, painting, and 3D design. So some San Francisco tech workers hooked ChatGPT up to Comma, an open-source system that’s normally used for self-driving but can also be used for drive-by-wire, which was plugged into a rented Toyota Corolla. A researcher sat in the car and fed camera and sensor signals from Comma into ChatGPT, which then decided on an action and sent speed and steering commands to the car. ChatGPT with GPT-6 Astra, OpenAI’s latest model, successfully completed a small obstacle course.

🚏🌃 The night sky is getting 10% brighter every year

A 2023 study involving thousands of citizen scientists found that the average night sky got 9.6% brighter every year from 2011 to 2022, which works out to a doubling in brightness every 7.5 years. In fact, astronomers believe that “over much of Earth’s land surface, the night sky no longer fully transitions to starlight and moonlight after sunset”: an intermediate “artificial twilight” caused by skyglow takes over the atmosphere after sunset. Fortunately, many places are weighing dark-sky laws, such as a proposed New York State law that would require people to turn off unshielded outdoor lights after 11 p.m.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Confessions of an Unrepentant Slop Snob (Charity Majors) — Explores workplace situations where reading AI-generated content feels offensive (personal thoughts and feelings) versus fine (code, and to some extent artifacts). Argues that we need to rework our communication norms for when and where we agree to use AI in documents, reviews, tickets, emails, etc. that we share with coworkers.

• The Lightbulb Computer (Guillaume Ardaud) [Also on YouTube] — A cool design prototype for a bulb-shaped smart device that could listen to voice commands, identify gestures, and beam text or images on walls, tables, or other surfaces. The designer envisions this world of “ambient” computing as a salutary alternative to recent trends like VR goggles that are “antisocial” and take you out of the moment.

• Jurassic Generation (The New Atlantis) — Examines the various themes of Jurassic Park, focusing on the illusion of control that science brings and the “unpredictability of tinkering with unfathomably complex natural systems.” When we see people “playing God,” as some fear about biotechnology and perhaps AI, it’s hard to point to a concrete danger, just the risk of unintended consequences — but Jurassic Park is a useful cultural touchstone that helps communicate the real perils.

• Patrick Gunkel Is an Idea Man Who Thinks in Lists (The Wall Street Journal) — A 1987 article that seems prescient: a quirky futurist named Patrick Gunkel developed a field called ideonomy, which crosses vast taxonomies to escape habitual thought, seeking unexpected connections by colliding unrelated ideas. It feels newly actionable with LLMs, which can interpret the combinations and help filter a hundred strange ideas down to the one worth pursuing.

🔮📬 Postcard from the Future

A ‘what if’ piece of speculative fiction about a possible future that could result from the systemic forces changing our world.

// What might teacher-parent conferences look like in 20 years?

Miss Brainmocker

To: Morgan-Smith@, Casey-Smith@

From: Teacher Rhodes

Subject: Parent-Teacher meeting setup

Hi, it’s time for the annual parent-teacher conference for 4th graders. I would like to chat with you about the rest of the school year ahead, as well as some behavioral issues I’ve seen.

<Agent items>

• Negotiation Items

  • Time / Date / Duration (15 minute minimum)

  • Device usage during meeting (type, duration, frequency, local bandwidth)

  • Recording / Transcription permissions

  • Parental Topics (open)

• Teacher agentic API [0x32….]

• Agentic permission level: Alpha-Three-Green

</Agent items>

// Several weeks later

The Smiths, a well-dressed couple, walk into a classroom. Teacher Rhodes stands, greets them, and gestures to some seats.

“Thank you so much for coming and for being actively involved in your child’s education. I see we’ve got green lights for protocols. I’ve reviewed the topics you wished to discuss. For the information items, I’ll beam over the data to your agents now. Let me know if you or your agents have any questions.”

The parents’ eyes defocus for a moment, as they review their agents’ take on the material. They nod at the teacher, no immediate need for in-person discussion.

“As for the behavioral issues I mentioned in my email, they revolve around your child’s agentic usage during school. As you know, 4th grade is when we introduce agents into the classroom, so that children can begin acclimating to their thought partners. The issue that I’ve been seeing is…”

A light chime rings and a soft voice says “agentic time enabled”. All three adults seamlessly shift to their agentic devices. Fingers and eyes twitch as they issue judgments to agents working on their behalf, everything from social scheduling to news consumption to personal passion projects to professional efforts. Ninety seconds pass, and the chime rings again.

The teacher’s eyes refocus on the parents. “The issue I’ve been seeing is that your child is having difficulty in some of the less basic usage patterns. Their swarm management, even with single digits of swarm units, suffers from a lack of oversight and delegation. Your child wants to focus on one topic, often for dozens of minutes at a time. This is not hyperfocus, but the undeveloped mental muscle of managing multiple agentic workstreams and retaining sufficient context on each.

“As an at-home exercise, have you considered introducing your child to StarCraft?”

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 252 — September 24th, 2026 — Available at read.fluxcollective.org/p/252

Contributors to this issue: Justin Quimby, Erika Rice Scherpelz, Neel Mehta, Boris Smus, MK

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Laura Eberly, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“The mind was dreaming. The world was its dream.”

— Jorge Luis Borges

🔎🔨 Forced lens swaps

At FLUX, we write a lot about Lenses. “Lens of the week” is one of our regular features! We are often asked why we focus so much on Lenses. Well, all views are limited. There’s no such thing as a “true” view of reality; there’s no view from nowhere. Sometimes consciously, but usually not, our views are shaped by our Lenses. Individuals have Lenses, and so do organizations.

After all, what is the innovator’s dilemma but an exploration of different Lenses newcomers and incumbents bring?

Whether individual or organizational, when we look through Lenses, we perceive the same situation in different ways, sometimes radically so. These Lenses may come from formal education. They may come from religious or cultural processes. They may come from organizational principles (both stated and implied) and individual lived experiences.

Regardless of origin, we all see the world through Lenses. We also see ourselves and others through Lenses; the actor–observer asymmetry is the meta-lens of why “I” am misunderstood and deserve grace but “they” are a jerk and deserve scorn.

Now, that’s all well and good, but still, why do we focus so much on Lenses? Well, when we become aware of our own Lenses, we can make a conscious choice to change the Lenses we are using. And naming a lens allows us to recognize it. It allows us to consciously take it on and off, to examine it and understand where it’s best used (and best set aside).

But we do not always have the luxury of choosing to change our Lenses. Sometimes, change is forced upon us or our organizations. This often takes the form of ground-truthing. The overwhelming accumulation of evidence and new facts shows the cracks in a currently used Lens. Kodak eventually had to accept that digital camera sales disproved the idea that its future lay in film. The machine guns and artillery barrages of World War 1 changed the worldview of European generals who expected short wars won by bold offensives.

The voluntary changing of Lenses can be a playful activity. However, forced Lens swaps often involve a massive and sustained avalanche of concrete, yet contrary facts. The more deeply entrenched the Lens, the deeper the blow to ego and self-esteem (individually and/or collectively) when we’re forced to abandon it.

In today’s world of shifting global power, LLM-mania, improving medicines, and changing climates, what Lenses might we be forced to change?

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏⚔️ “Intelligence” from an AI chatbot almost started a US–China military incident

An intelligence report circulated this spring, during the war with Iran, “set off alarm bells” for the US military when it ‘found’ that a Chinese ship in the Middle East was “transporting components of a nuclear weapons program.” The American military had planes in the air and armed personnel were reportedly preparing to forcibly board the ship. But when officials looked more closely, they realized that the report had been made by an analyst who used a generative AI chatbot, and the report itself was “entirely false.” The AI had apparently misidentified the materials the ship was carrying, and yet, according to a source, it had “almost started a war.”

🚏🛬 China took stealth aircraft parts after a plane was diverted to Hong Kong

An intermediary working on Lockheed Martin’s behalf was transporting components from an F-35 stealth fighter jet from Australia to the US for inspection and sent them via a UPS flight. But the plane unexpectedly stopped in South Korea and got diverted to Hong Kong, at which point the Chinese government seized the sensitive parts and “so far hasn’t returned them.” It’s unclear whether the diversion was accidental or intentional. (China has long shown interest in getting data and parts for the F-35, including through electronic espionage.)

🚏🎣 Newfoundland’s cod stock has recovered, 34 years after its infamous collapse

Decades of overfishing near Newfoundland made cod populations collapse by 99%, leading to the devastating 1992 moratorium on cod fishing in Atlantic Canada, which was the “largest industrial closure in Canadian history” and led over 30,000 people in Newfoundland and Labrador to lose their jobs. Many towns were abandoned, and the economy was depressed for decades. But a generation later, cod populations have rebounded, and NL’s cod stock has been deemed “healthy” for the first time since ‘92. The moratorium was lifted two years ago (with a quota that was raised by 55% this year), and fisheries and processing plants are hiring and expanding again.

🚏🧑‍⚖️ A ChatGPT-using lawyer cited fake quotes from fake witnesses

In a recent criminal case, a defense attorney from New Mexico used ChatGPT to generate a legal brief that included “false testimony from wholly fabricated witnesses,” fabricated quotes from real witnesses, and “inaccurately described [legal] precedents.” He admitted to feeding the trial transcript (which was itself generated by an AI transcription service) into ChatGPT, after which it outputted the fake quotes. The judge raked the lawyer over the coals, hitting him with a fine and referring him to a disciplinary board. In a statement, the lawyer said he “didn’t know that AI could hallucinate facts” and asked the disciplinary board to take that into account. This marks yet another entry into the common genre of lawyers getting busted for using AI to generate fake information.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Do You Know Your Default Shape? (Useful Fictions) — Presents a theory that many people (everyone?) tend to have a default lens through which they initially see the world, whether it’s marriage or politics. It’s a cogent reminder to learn your default lens but try to hold it more loosely, since even the most useful frame eventually blocks light you need.

• The Road to Epsilon-Zero: Ordinals as Nim-Heaps (The Universe of Disclosure) — A mathematician uses a children’s game about moving beans to spin an impressively accessible introduction to ordinals, which in set theory are extensions of natural numbers to varying levels of infinity. Read the whole (in-progress) series for such gems as, “in at most 17 moves, we will be able to say, not how long the game will take to finish, but how long it will be before we can say how long the game will take to finish.”

• Why Buffets Suddenly Disappeared (Modern MBA / YouTube) — Argues that all-you-can-eat buffet chains, a staple of the ‘80s and ‘90s, were a product of a peculiar time in American history where incomes were low and land, labor, and food were cheap. Their inexorable decline in the 2000s was an interesting case study for how no number of margin-boosting tweaks can compensate for declining demand in the category, and how cash-cow businesses in managed decline are a perfect target for private equity.

• Meet the Sad Wives of AI (Wired) — Examines the domestic strain of the AI gold rush, where the obsessive, all-consuming race to build alienates partners and hollows out local social circles. Though the piece leans heavily on gendered tropes, the underlying dynamic is a classic “pioneer” FOMO: the crushing panic that stepping away from the machine for even five minutes means losing your edge.

🔍📆 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: fit for purpose.

The Mercator projection is infamous for distorting the real sizes of landmasses. It introduces size distortions that are quite extreme the further a point is from the equator. It’s a bad map, right? Well, it depends on what you’re using it for. If you’re a sailor trying to chart a constant bearing, it can be quite valuable. Whether or not a map is “correct” depends on what you’re using it for.

Maps are a particularly clear example of being fit for purpose. An object is fit for purpose if it works well for the job it is put to. This feels obvious, but in practice, it can be anything but.

Sometimes an object is not fit for purpose because it’s the wrong tool for the job. Maybe it’s, unironically, 10,000 spoons when all you need is a knife. Other times, the right tool does not exist. Lego, founded in 1932, introduced its modern brick in 1958, but the brick separator wasn’t introduced until 1990. Sometimes a tool is not fit for purpose because it’s trying to be too much. It’s the 3-in-1 shampoo+conditioner+bodywash that doesn’t really do any of its jobs well.

But see what we did there? We assumed a purpose! If our purpose is to save space in our luggage, perhaps the 3-in-1 is exactly what we need. The deeper insight of fit for purpose isn’t that some things are better than others, but that the purpose of these things is plural.

So the next time you find yourself wondering why something is the way it is, try to see if there is some purpose it may be fit for… even if that purpose is a far cry from yours.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 251 — September 17th, 2026 — Available at read.fluxcollective.org/p/251

Contributors to this issue: Neel Mehta, Boris Smus, Ade Oshineye, Ben Mathes, Dart Lindsley, Erika Rice Scherpelz, Jon Lebensold, Stefano Mazzocchi, MK

Additional insights from: Jasen Robillard, Justin Quimby, Laura Eberly, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“Fast learns, slow remembers.”

— Stewart Brand, The Clock of the Long Now (1999)

🫟🍞 Tuning the rate of change

Heat boils water. Kneading transforms goopy flour into silky dough. Conflict brings an organization suddenly together… or tears it apart. These are all phase changes. The system shifts in a way that fundamentally impacts its nature. Sometimes (as in the case of water) only as long as the external condition applies. Sometimes (as in the kneading) irreversibly.

In human systems, it can be hard to tell what changes are tied to the external pressure and which get baked into the system. For example, neither remote work nor mask-wearing was as persistent after COVID-19 as some thought they would be. Vaccine technology, on the other hand, was changed forever.

Phase transitions can feel like they have a clear shape. An external force comes and it causes change. Water boils at 100°C. But these “absolutes” can change. Physical changes such as adding salt or adjusting air pressure can change the temperature at which water boils. Chemical catalysts, such as digestive enzymes that break down food more quickly, can speed up a reaction. Kneading time changes the nature of dough, and so does the strength of flour you use.

What about in organizations? One “catalyst” for change in organizations is organizational maturity. The maturity of the team, the maturity of the technology, the size of the organization, all of these come together to impact how phase shifts happen. You could look at maturity as one type of agent (in the chemical reaction sense): adding it to the mix changes the phase transition point. We can’t change the maturity of an organization on a dime, but if we’re looking for organizations ripe for change, we can take it into account.

Other change agents can be applied from the inside. Glue work, support functions, change management can all act as catalysts for phase changes. These often-underappreciated factors (and people) provide the social lubrication that reduces the friction produced by organizational headwinds… or sometimes they provide the friction.

An important thing to remember about catalysts is that they cannot cause change on their own. Catalysts only speed up a reaction that’s already happening. The conditions need to be in place to trigger a phase change in the first place.

And speed isn’t always the desired outcome. If a change is happening too quickly, sometimes you need inhibitors, factors that slow down a reaction without stopping it. Inhibitors acknowledge that sometimes a change is good… but the surrounding system’s ability to absorb it must be taken into account to avoid adverse side effects.

So when a system undergoes change, think not only about outcome but also about the rate of change and how to accelerate or moderate it to be most effective. Maybe it kneads a little more pushing and pulling to really come out smoothly.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🇦🇺 EV sales overtook petrol car sales in Australia

August marked the first month in history that more battery-electric vehicles were sold in Australia (27,089) than fully petrol-powered vehicles (25,824); hybrids are counted separately and also account for a large chunk of sales. It will still take a while for EVs to dominate Aussie streets: they currently make up just 3% of cars on the road, but by 2050 they’re projected to represent 97% of all light cars down under.

🚏🍿 AI-assisted ‘micro dramas’ are booming in China

While Hollywood wrestles with how to incorporate AI, Chinese film studios are leaning heavily into it through the genre of micro dramas, which are one- to two-minute, movie-style videos that are “devoured by Chinese audiences” on apps like Douyin. About 128,000 such mini-films were made in the first quarter of this year alone, and over 95% of them were made with AI; they’re a good fit for AI because their disposable nature means they need less production value or skilled acting. On the other end of the spectrum, Chinese movie studios are creating 90-minute feature films with AI for about $500,000, compared to budgets deep in the millions for conventionally produced movies.

🚏🇺🇸 Dozens of congressional candidates are paying for ChatGPT, even where it’s not allowed

An analysis of campaign finance disclosures found that 39 candidates for the US House and Senate (whose midterm elections are under two months away) reported paying for an OpenAI subscription. Two candidates revealed they were using ChatGPT to generate ads, even though OpenAI bans that practice for political candidates; another candidate disclosed that they were using AI to “draft and personalize messages or create synthetic media.” The top-spending candidate has spent around $3,260 on ChatGPT since 2025, so it’s not a major expenditure in the grand scheme of things. (Campaign consultants, political PR and advertising professionals, and PACs may be spending a lot more on AI, though.)

🚏😎 California got through a brutal heatwave with no rolling blackouts

Back in 2020, California suffered a massive heatwave that triggered rolling blackouts due to excessive cooling needs. Last week, the Golden State melted under another record-breaking heatwave, but it didn’t have a single rolling blackout. The reason: California has grown its grid’s battery storage capacity more than 30x in the last six years, which helps the grid handle sudden spikes in power consumption while squirreling away extra power (largely from solar farms) when there’s a surplus. It also helps that California’s grid is now integrated with that of the Pacific Northwest, so California can pull power from its cooler northern neighbors when heatwaves strike.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• On Average (Monkey’s Uncle) — A Stanford anthropologist observes that maximizing theoretical expected value can, in reality, lead to ruin. One reason is using the wrong type of average: arithmetic means (which power expected value calculations) are fine for population-level analysis, but if you’re repeatedly playing a game you have to consider the geometric mean instead. There are also “absorbing boundaries,” such as going broke when gambling, which make simple EV math inapplicable. This is the world of ergodicity economics, which we enjoy discussing here at FLUX.

• Dude, You Broke the Future! (antipope.org) — Sci-fi author Charlie Stross inverts the AI timeline: the dangerous artificial intelligence isn’t coming, it arrived in the 1500s. Corporations were humanity’s first AIs: slow, legally recognized artificial persons maximizing not paperclips, but revenue.

• Scorched Harvest (4Hunger.org) — A detailed projection of the dire climatic and agricultural impacts of this year’s El Niño, forecast to have a good chance of becoming the strongest on record. Irregular patterns of floods and droughts could cause crop failures across the world, from Australian wheat to Indian rice to southern African maize, and if history is any guide, it could lead to famine and inflation through 2027 and 2028, plus all manner of unexpected consequences. (El Niño is a strange phenomenon in that we can predict it very well, unlike random storms, but its impact is no less brutal).

• How Many Cities Should Be in Middle-Earth? (Improbable Matter / YouTube) — Uses probability distribution functions and England’s oldest census to predict the distribution of city populations in rural and urbanized areas. Then, applies these findings to fantasy worldbuilding to predict the kinds of cities we’d expect to see in Tolkien’s Middle Earth.

🔍📆 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: resolution asymmetry.

If social media is to be trusted (it isn’t), then university campuses are battlegrounds where civil discourse has died. Walk onto an actual quad at noon and you’ll see freshmen on hammocks, furious debates over mathematical platonism, and raging over… the broken waffle iron in the dining hall. The controversies and trigger warnings, debates about wokeness, cancel culture, and the like may be real, but they are only a minor part of the lived experience.

We can call this resolution asymmetry. There is a gap between the high-dimensional experience of living in a system and the flattened, low-resolution view from the outside. Observing a system from a distance requires compression, often in the form of low-bandwidth proxies such as viral headlines, limited dashboards, or nostalgic memory.

Real systems are almost always full of contradictory signals. But from the outside, these signals are all collapsed into a low-dimensional vector. Is this place thriving or ruined? Is that government pure or corrupt? Someone who has never been inside can mistake the caricature for the territory. For those on the inside… few things are as surreal as having a stranger confidently (and wrongly) explain your own daily life to you.

Even if we know better, we can fall into this trap. So, where might you be relying on a flattened view of a system you don’t inhabit? And what texture gets lost when others view your world from the outside?

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 250 — September 10th, 2026 — Available at read.fluxcollective.org/p/250

Contributors to this issue: Neel Mehta, Boris Smus, Justin Quimby, Ben Mathes, Ade Oshineye, Erika Rice Scherpelz,

Additional insights from: Dart Lindsley, Jasen Robillard, Jon Lebensold, Lisie Lillianfeld, MK, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“To live effectively is to live with adequate information.”

— Norbert Wiener, The Human Use of Human Beings (1950)

📝 Editor’s note: Thanks for supporting us through five-plus years and 250 episodes!

〰️☂️ Leading through the fog of work

Should the onboarding module start with basic functionality or our app’s secret sauce? It’s a legitimate debate. But what the team doesn’t know: the debate may be irrelevant. Upstairs, leadership is discussing how our biggest renewal fell through… and we may not even survive the next six months unless we pivot and focus on our small, profitable core. The team’s manager is in both discussions, but they don’t give even a hint that the team should focus on higher priority tasks. They don’t want to cause churn when there’s a chance that something may rescue the status quo.

From a place of genuine care, the leader is acting as a “shit umbrella.” They don’t want to distract the team with anxiety-inducing volatility and churn. This is fine up to a point, but when an event finally gets through the umbrella, the team may be confronted by months of accumulated drift from the company’s reality. This kind of top-down alteration of signals is akin to the ‘watermelon status’ lens that we discussed back in issue 36. It’s also why the vast majority of shit umbrellas eventually become shit funnels. Over time, the disconnect with reality catches up to them or the manager burns out, leaving the team exposed to a reality they don’t know how to handle.

That’s why this week we want to look at another metaphor to help teams ride the waves of change: the low-pass filter. It blocks rapid, spiky, volatile high-frequency signals, like a rumor before a board meeting or hubbub over a competitor’s blog post. The more consequential low-frequency signals pass to the team, such as a change in the sales team’s Q3 focus or falling revenues.

An umbrella blocks both useful and harmful signals, but a leader who provides no filtering subjects the team to whiplash. However, the low-pass filter aims to dampen the jitter but preserve the trend. Not every competitor announcement matters… but if competitors are consistently winning out, that’s a signal to pay attention to. The communication challenge is avoiding either extreme.

What does this look like in practice? It can mean acting as an interpreter, communicating synthesized information rather than the play-by-play. Or it can mean a leader can readjust plans at a regular cadence but keep stability within a cycle so that when communicating about the bumps, we provide a broader perspective.

The low-pass filter isn’t the only, or always the best, response to a shifting environment. For example, a leader can work to change the environment so that transparency doesn’t feel unsafe.

Protecting a team from noise is an act of stewardship. But shielding them from reality leaves them unprepared for the terrain ahead. The low-pass filter gives us a tool for preparing the team over time to be able to handle the sometimes-harsh reality of your organizational environment.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🚢 A Panama Canal transit slot went for nearly 100x its normal price

Ships that arrive at the Panama Canal without a reserved crossing time can bid for available slots as part of its transit auction; the average winning price (before this year) was about $55,000. But in August, a South Korean gas ship paid a record $5.3 million for a transit slot, beating the previous record auction price of $4.6 million set earlier that month. The disruptions in the Strait of Hormuz and Bab el-Mandeb likely redirected much traffic flow to Panama. Indeed, the share of vessels approaching the Canal without a booking rose from 10% to 20–25%, and the number of daily bidders rose from “two to three” to a peak of eighteen, which reflects the volume of last-minute changes of plan and the lack of easy alternatives.

🚏🍷 Italy saw record-early grape harvests amid extreme heat

This summer’s heatwaves in Italy forced winemakers to start harvesting grapes earlier than ever before: growers in the Oltrepò Pavese, in the northern province of Pavia, had to start their pinot noir harvest on July 27th, compared to the usual start date of August 10th to 15th. Similarly, winemakers in Lombardy started harvesting the famous grapes used in Franciacorta sparkling wine ten days earlier than last year, and “about a month earlier than three decades ago.” High heat makes grapes ripen more quickly, which raises sugar content and reduces acidity, so grape growers have to be very careful about harvesting on time. Indeed, many in the industry believe that global warming is pushing wine to higher latitudes, such as Britain, Tasmania, Washington State, and Patagonia; up to 90% of traditional wine regions in the Mediterranean and California “could be at risk of disappearing by the end of this century.”

🚏🪨 A Go grandmaster beat the top AI engine (with a two-stone head start)

A decade after the AI program AlphaGo famously defeated the top human player, another grandmaster beat the world’s new top AI Go engine, KataGo, by two matches to one. The grandmaster, Shin Jin-seo, dropped the first match but observed he had to play defensively and preserve territory at the start, only pivoting to counterattack in the middlegame, which helped him win the last two games. Shin later reflected that copying the AI’s strategy didn’t work, and instead he had to “build the board according to [his] own style.” (Interestingly, this mirrors chess, where the unusual, defensive Hippopotamus opening can hold its own against top engines, whereas aggressive gambits struggle.)

🚏🍎 NYC is banning AI for students up to 8th grade

New York City mayor Zohran Mamdani announced that the city is placing a moratorium on generative AI use for public school students between 2-K (a preschool program for two-year-olds) and 8th grade; high school students can use AI but will get twice-annual AI literacy classes. The city says it’s the “most expansive AI moratorium in the nation.” NYC’s public school district, the largest in the USA, previously banned phones, tablets, and smartwatches throughout the school day for all students, including high schoolers.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Why ‘Cost Disease’ Is the Secret Force Behind America’s Toxic Solitude (Derek Thompson) — Argues that Baumol’s cost disease is the invisible force hollowing out our civic infrastructure: while digital, hyper-scalable platforms like Netflix get cheaper and more productive, labor-intensive third places like cafés and independent bookstores become unsustainably expensive. This economic gravity favors solitude because, unlike the physical world, it is highly scalable. Thompson’s antidote is a structural shift toward sustainable models of conviviality such as membership-funded third places and paid civic clubs to consciously finance and protect human-scale connection.

• AI Is Technology, Not a Product (John Gruber) — Argues that the popular line that “Apple needs to launch a killer AI product” is nonsensical, because no matter how good personal AI agents get, they will always need to be mediated through some device, and nobody can build small personal computing devices better than Apple. Apple will need to upgrade its product suite to incorporate AI, just as it does with every new technology, but for Apple, technologies are just a means to an end.

• “Hybrid War” Is a Red Herring (Kiran Pfitzner) — A Clausewitz scholar argues that the popular concept of “hybrid war” (where cyberattacks, misinformation, and plausibly deniable attacks like Russia’s ‘little green men’ augment conventional military actions) is a category error and obscures more than it explains. The important thing in war isn’t the specific tactics but rather the political situations that make said tactics work. For instance, the useful insight about the ‘little green men’ wasn’t the action in itself but rather that Russia’s 2014 campaign took place at a time when the West wasn’t interested in intervening in Eastern Europe, and the ‘little green men’ gave the West a convenient excuse to ignore the move.

• How 9/11 Shaped My Life (TIME Magazine) — Twenty-five Americans from New York and beyond, spanning firefighters, lawyers, journalists, immigrants, and more, reflect on how that fall morning a quarter-century ago changed their lives.

🧩🕹️ Systemic tools and toys

Games, gadgets, demos, and other nifty interactives to stretch your systems-thinking capacity.

American English is regarded as fairly homogeneous, especially among younger and more educated speakers, but dialectal words and slight pronunciation differences can still identify speakers’ regional accents. The new web quiz “How Do You Talk?” asks targeted questions to guess which part of the USA your speech most resembles, drawing on a famous dialect dataset. If you’re an American English speaker, try out the short, normal, or extended quizzes and see if they get your origin right!

Here are one Philadelphia-born FLUXer’s results. Very close, although the “hoagie” question made it obvious. (Go Birds!)

🔮📬 Postcard from the future

A ‘what if’ piece of speculative fiction about a possible future that could result from the systemic forces changing our world.

// 2074. A college lecture hall.

“Alright, Sam, you are the lucky student today. Please come down to the front of class. While they are doing that, here’s a reminder of what we are doing.”

“You are at this University because you want to learn. Not just recite dates and names, but also actively discuss those topics, not just cut and paste a response from an LLM agent. Each week one of you starts the class by talking about the topic from the syllabus. Not only does this get you to know the material, but you will also get to practice public speaking and thinking on your feet, both critical skills in today’s world.”

// Student arrives at the front of the class and sits on a swivel chair across from the professor

“Alright, please explain what drove the ‘Financial Nihilism’ of the early 2030s.”

“Financial Nihilism is a term that rose in prominence in the late 2020s as the US took a hard and visible K shape. Bond rates skyrocketed across the G7 after the Russian and Hormuz oil shock years combined with AI’s white collar unemployment outside the major metros, causing the US market to crash in 2028. The subsequent chaos led to significant rage among Gen Z and Gen Alpha. With the doorway to home ownership seemingly shut thanks to high unemployment and high interest rates, online prediction markets and gambling became more and more common for the young as a way to try to escape their sense that they had no way to get a start on their life or career.

“It was in the 2030s that the Financial Nihilism movement transitioned from a purely economic sentiment to breaching into the real world. Online rage transformed into real-world violence. “Takeovers” of rich gated communities became a regular phenomenon. Despite institutional investors only owning about 3-5% of US housing stock, populists wanted someone to be mad at for their housing costs and set fire to the headquarters of the two largest residential real-estate holders. Attacks by unregistered drones rose against data centers and bank headquarters. Then when the baby boomers began dying of old age in waves, the housing market experienced a flood of homes entering the market. Normally this would lower prices, but the individuals who inherited the homes didn’t want to recognize a lower home value, so housing supply paradoxically shriveled even more. Each homeowner was like Wile E. Coyote, running off the edge of a cliff hoping they’d reach the other side of a canyon before they looked down.”

The professor rises and shakes the student’s hand. “Well done! Now take your seat while we explore how Financial Nihilism impacted federal and state tax codes in the late 2030s…”

Choose Your Own Adventure:

- for optimistic radicalism, turn to page 37 for The Land Value Tax

- for trying ideas already proven to backfire, turn to page 47 for The Wealth Tax

- for war, pestilence, and destruction, turn to page 66 for The Great Leveler

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 249 — August 27th, 2026 — Available at read.fluxcollective.org/p/249

Contributors to this issue: Laura Eberly, Justin Quimby, Erika Rice Scherpelz, Neel Mehta, Boris Smus, MK

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“We dream, we wake on a cold hillside, we pursue the dream again. In the beginning was the dream, and the work of disenchantment never ends.”

— Kim Stanley Robinson

📝 Editor’s note: We’ll be off next week for the US Labor Day holiday. We’ll see you again the week after!

🖊️🤖 Language evolves, truth-telling endures

What’s a writer to do? Some of us have been using em-dashes since before they were uncool. Emojis to start headings are a long FLUX tradition. And we’ve even been known to delve on occasion. Yet nowadays, all of these are “tells” for AI writing. People who never noticed these patterns before are suddenly greatly concerned about their use. There’s pressure to remove AI-associated language, lest the writing be perceived as inauthentic. However, LLMs are trained on vast amounts of textual data. If these patterns appear in AI-assisted writing, it’s because they appeared in human writing first.

A piece of the puzzle is that superficial writing quality used to be “proof of thought.” Similar to how Latin was once the lingua franca of scholars, the proper usage of an em-dash or less common vocabulary correlated with someone who had put thought into what they were writing. “Smart” writing had a veneer of intelligence, signaling a certain type of education, whether or not the underlying idea deserved it.

Now these are decoupled. It’s easier than ever for a mediocre idea to hide behind writing that passes the sniff test of having been reasoned out. The easy-to-detect markers are no longer reliable. And because AI-generated writing often uses these markers, we have started to become suspicious whenever we see them.

In a recent example, Hank Green was “caught” using AI because his script included the phrase “Thanks for the pushback.” Why the scare quotes? People thought they found a “tell.” It turned out that the phrase was not generated by AI, although he admitted to using AI elsewhere in his creative process in ways that he felt were unhealthy. A coincidence became a scandal because people thought they could sniff out “just paste the AI” writing and use it to make judgments about the entirety of Hank’s work. Even the discussion gets meta. In a long-form, multi-generational discussion on Hank Green’s AI use, one of the commentators said — unscripted, unironically — “I appreciate the push.” They then joked that the speaker was “secretly AI” who could no longer be trusted.

Are LLM “tells” erasing legitimate use of language? LLMs first learned those patterns from good writing. Those LLMs are then used in generating writing both good and bad (witness the explosive growth of slop on LinkedIn). Now those patterns have a negative association. One way this goes is that parts of language are lost to humans, given over to AI entirely. Then, when good writing shifts and LLMs go through a future round of training, new phrases are denylisted.

Or we can go another direction. As readers and writers we can hold ourselves accountable to ideas. Sometimes we need to delve into an idea — perhaps with a side thought here and there — regardless of what the AI detector thinks 😉. But don’t stop there. AI writing may appear good at first glance, but upon closer inspection, the writing is often internally incoherent. As writers, we can ask ourselves, “What am I really trying to say? Did I accomplish that?” As readers, we can ask ourselves, “What is the underlying idea here?” And for lower-stakes writing, sometimes all it takes is a round of reading and editing.

Writing quality does matter. Smart-sounding writing getting cheaper will change what we perceive as good or bad. However, at the end of the day, the truth of a piece is what will last, no matter how it’s expressed.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🎯 Canada’s retaliatory tariffs are targeting states with upcoming Senate elections

After US President Donald Trump slapped 50% tariffs on a wide range of Canadian imports this week, Canadian Prime Minister Mark Carney announced his country will unleash “dollar for dollar” retaliatory tariffs next month. Carney appears to be strategically picking industries to target swing states that have Senate elections this November: tariffs on steel and auto parts will affect Michigan, metal tariffs will hit Ohio, seafood tariffs will hit Maine and Alaska, and agriculture tariffs will affect Maine and Iowa. Meanwhile, Ontario Premier Doug Ford wrote a letter to Carney advising him to focus tariffs on imports from “politically significant states to the current administration’s base of support,” naming eight mostly Republican states: Alabama, Arkansas, Florida, Iowa, Missouri, Montana, Texas and Wisconsin.

🚏🚰 Iranian hackers targeted over 100 water systems across the USA

Last month, a spate of cyberattacks from Iranian hackers shut down water plants in several large suburbs in Minnesota, marking a new front in the US-Iranian war. It turns out that wasn’t an isolated incident: the Cybersecurity and Infrastructure Security Agency (CISA) issued a report this week that revealed that “malicious cyber activity” had targeted “over 100 internet-exposed [water] systems” nationwide, primarily through internet-connected programmable logic controllers (rugged computers designed for industrial settings). Agencies said that hackers are using “AI-generated exploitation scripts,” and cybersecurity experts warn that the numerous attacks so far may be “test runs for a larger-scale attack.”

🚏🌞 American solar generation is up 20% year-over-year

Although coal and natural gas plants have been making a heavily publicized comeback in the US, solar power is continuing to accelerate regardless. New data from the Energy Information Administration shows that solar generation hit a record high and is up 20% compared to this time last year. Solar now represents 9.3% of US electricity generation (averaged across the last twelve months), another record high. Texas has, somewhat surprisingly, been one of the top solar states: it saw a 31% year-over-year increase in solar production, and solar now makes up 12.2% of the Lone Star State’s electricity production.

🚏♨️ The world’s oceans hit their highest temperature on record during El Niño

The highest sea surface temperatures typically appear in March or April, toward the tail end of summer in the watery southern hemisphere, and yet this August 22nd, in the depths of the southern winter, set an all-time record for highest sea surface temperature at 21.1 °C (70 °F). This year’s unusually strong El Niño is partly to blame, as is global warming, which stores far more heat in the oceans than in the air.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• SaaS Isn’t Dead. Sameness Is. (Chad Fowler) — Argues that the foundational assumption of software-as-a-service – that thousands of customers could all share the same software – is broken now that custom apps and interfaces are easier to generate than ever. Still, there’s an advantage to everyone sharing the same underlying infrastructure (identity, security, data models, payment rails, etc). The author likens it to pace layers: everyone will build their own applications (the faster-moving layers) on top of a shared, stable substrate (the slower-moving layers). SaaS companies will survive and thrive if they become substrate companies and differentiate based on their infra, not their interface.

• Why the West Can’t Build a Cheap EV (Modern MBA) — Argues that the “outsource everything” business model of America’s legacy automakers left them structurally unable to win in the EV market, where vertical integration is key. This is where Chinese upstarts have come to dominate the EV space. It also helps that the Chinese government invested early and often in the industry through charging networks, tax breaks, policy concessions, and more, treating EVs as a strategic national priority. Meanwhile, the US showers benefits on already successful for-profit companies, which is great for innovation and shareholder returns but struggles with long-term, capital-intensive infrastructure projects.

• Dependence Depends, Doesn’t It? (Dan Davies) — Argues that, in international relations, weaponized interdependence (controlling an important chokepoint that everyone needs) works better as a threat than as something you actually use. Systems are pretty good at routing around chokepoints, so if you block one and everyone adapts, you’re revealed to be a paper tiger. Plus, sometimes blocking the chokepoint hurts you more than the enemy. Better to bluff than to let things reach a showdown.

• The Megaflood That Created the Mediterranean (Uncharted Territories) — Recounts how, 5.3 million years ago, the fully dried-out Mediterranean basin refilled in mere months via a 1.5 km tall waterfall, with inflows over 100 times that of all the world’s rivers combined. Given the scale of catastrophe it’s tempting (but completely wrong) to read this Zanclean flood as the source of the flood myths present in many human cultures, as it predates humans by millions of years.

🔮📬 Postcard from the Future

A ‘what if’ piece of speculative fiction about a possible future that could result from the systemic forces changing our world.

// How might people try to escape the “AI hiring doom loop” in the coming decade?

// 2053. A cozy family room in a tasteful home, windows overlooking a wooded valley. The wall opposite the windows is full of books and the occasional industry award. A mug of tea sits beside the rocking chair, occupied by the woman who won those awards. Her niece sits on the couch, a frustrated yet inquisitive look on her face.

“I don’t get it, Auntie! In our ‘History of the 21st century’ class, I’ve been learning all about how the early phases of AI ricocheted through the economy in the late 20s before the Big AI Retraction of the 30s. From everything I’ve read and seen, it sounds like getting a job in that era was really hard. So how did you manage to get a job at one of the top companies at that time? We played through some simulations of what job hunting was like with the tools available and not a single person in my pod was able to even land a callback!”

The older lady smiles and chortles before taking a sip of her tea.

“Oh, that was a rough time. Fresh grads used AI to apply for jobs at scale with perfectly targeted resumes and cover letters. Companies, facing 1,000x the regular scale of applicants, turned to AI to filter applicants. It was an AI hiring arms race, with each person’s career on the line.

“In-person networks and referrals became the fallback, which then led to candidates stalking hiring managers, or anyone else. The front door to jobs was jammed, so people tried every side door they could. I looked at the problem slightly differently.”

She reaches to the table her tea mug sits on and flips a small switch. A light hum emanates from the bookcase. The niece’s eyes go wide. “What happened to my BrainPal? I can’t connect to the netcloud!?!”

The Aunt pats the base of the bookcase. “Just a small EM field to jam transmissions. I don’t want this part of our conversation getting added to the training data set just yet!

“Where was I? Oh right. So for me, I didn’t want to try the side door, I went for the Back Door. And by that, I mean I took the AI tools available and used them to find security weaknesses in every company I wanted to get hired at. And the playbook was relatively simple: first, penetrate their network security. Second, figure out which AI tools and systems they were using for the hiring process. Third, inject data into the company-specific hiring data which would boost my recommendation score and downgrade other applicants at each stage of the hiring pipeline. I never put myself at the top, but in the top three. That way, I’d get to the in-person interviews. At the end of the hiring process, hired or not, I’d then erase my tracks.

“When faced with the Kobayashi Maru, it pays to think differently.”

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 248 — August 20th, 2026 — Available at read.fluxcollective.org/p/248

Contributors to this issue: Neel Mehta, Boris Smus, MK, Erika Rice Scherpelz

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Justin Quimby, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“The only thing that makes life possible is permanent, intolerable uncertainty.”

— Urusla K. LeGuin

🌀🔵 The shape of confusion

The chaos of a roomful of preschoolers can be overwhelming. It can also be a place of immense creativity… admittedly often of the “what can go up a nose” variety.

But what do we mean by “chaos”? In the colloquial sense, chaos means disorder. But in the mathematical sense, which Cynefin pulls from, it means a system that is unpredictable because it’s highly sensitive to initial conditions, one where dramatically different outcomes sit quite close to one another in the outcome space. Consequently, chaos contains patterns that are unstable and often only visible at the macro level.

Chaos often requires action, observation, and response rather than predictive planning. Those preschoolers look like the peak of disorder at first, but as you observe them, you can see that there is a system to their actions, an act-sense-response pattern that is representative of the Cynefin chaotic domain. This is what makes it creative.

It is also what makes it dangerous.

The realm of confusion is interesting because it sits in the center of the Cynefin diagram and yet is often left out completely in conversation and consideration. This is because confusion is a bit different from other domains: while the clear, complicated, complex, and even chaotic domains all reflect how we describe the problem itself, confusion is about our state of knowledge of the underlying problem.

Confusion is the realm of Eris, the goddess of discord and strife. One person may see a problem as simple because their analysis assumes a cause-and-effect relationship while another may see it as a chaotic mess due to getting caught up in patterns they don’t have the models to comprehend. Or it’s a multitude of challenges with different properties that may all be tied up together, waiting to be separated. Understandably, confusion can breed disagreement.

But also, confusion opens up the possibility of change. It reveals that our understanding of the problem can change the shape of the problem. And once we see that possibility, we realize that we can change the nature of problems even when we think we know what shape they take.

For example, weather is the epitome of a chaotic system, but it can be reduced to what most individuals experience as a complicated system when we realize that we can collapse a broad swath of states into a temperature range and a probability of rain. (Or, perhaps, a simple system, if you’re in San Diego, where “about 70°F and sunny” is almost always the right answer. 😉) Another transformative move is the addition of rules or guardrails: a complex system of human interactions gets simplified into abstractions of teams and organizations.

Transformations can also go the other direction. The historically simple choice of what to eat — whatever we can get our hands on — moved to complicated as choices increase. And it can move to complex once we start accounting for individual dietary needs.

This doesn’t mean transformation is arbitrarily possible. We always have to check framing against the underlying reality. But it’s worth recognizing that chaos, like other states, has an element of perspective to it.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏📊 People in many countries now view China more positively than the US

Pew Research polled people from 36 countries and found that their opinions of China have steadily improved since 2023, while their feelings about the US have soured, to the point where China is now viewed more favorably than the US, on average. Similarly, people have more confidence in China than the US to “do the right thing regarding world affairs.” China’s favorability lead is strongest in South and Southeast Asia, while enjoying a modest lead in Europe and Africa, while the US retains a favorability edge in Israel, Japan, India, South Korea, and not many other places.

🚏🌡️ Solar panels helped Europe meet power demand during recent heat waves

During this summer’s heat waves, European electricity demand rose by about 25% due to increased A/C use. Low river levels meant that hydropower and even coal plants (which need water for cooling) had to reduce energy generation during the heat waves, while solar installations actually got 17% more productive during this span – the only power source to do “better than usual.” Increased solar radiation is probably to credit for the increased solar output, although high temperatures are known to slightly reduce solar panels’ efficiency.

🚏🦦 Otters were spotted in NYC’s Bronx River for the first time in 100 years

This summer, a river otter was captured on camera in the Bronx River, the only freshwater river in New York City, marking the first documented sighting of a river otter in this waterway since the early 1900s. The Bronx River was “treated as an open sewer and dumping ground” during NYC’s industrial boom in the 19th and 20th centuries, so the return of otters (a few decades after beavers returned after 200 years) suggests that long-running cleanup operations have been successful.

🚏🛬 AI companies bid millions of dollars for Spirit Airlines’ internal data

After Spirit Airlines went out of business a few months ago, Google paid $10 million for its internal data, including “email and internal communications, spreadsheets, transactions… as well as human resources information,” presumably to help train Google’s AI models. Mercor, a company that provides human expertise to AI labs, submitted the second-highest bid at $7.5 million. It’s an unusual outcome because defunct US airlines are typically sold wholesale – data, planes, and all – to other airlines, rather than ceasing operations and being sold off for parts.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Hard Is Not Defensible (Alex Crompton) — Differentiates between two kinds of “hard” in building startups: “math hard,” where innovations are hard to come up with but easy to copy, and “bodybuilding hard,” where everyone knows the method for success but few can actually stick it out. But either way, they don’t give you a sustainable moat; they just get you started and buy you time until you can find a real advantage, which relies on “unique resource accumulation.” Thus, the key to success: “you need to snowball.”

• What Really Happened on Easter Island? (Scientific American) — Argues that the popular story of ecocide – that the indigenous people of Rapa Nui destroyed their own environment, which led to their downfall – was actually a myth perpetrated by the pop-sci book Collapse, which was trying to weave a cautionary tale about preserving natural resources. In truth, Easter Islanders maintained a stable population and economy for centuries, using clever hydro-engineering and agriculture techniques to scratch out a living on an inhospitable rock. It was only after European colonizers arrived, along with diseases and economic exploitation, that the island fell into ecological ruin.

• The Last Useful Man (Metropolitan Review) — A review of the final installment of Mission Impossible, where the heroes are fundamentally useful and self-sufficient, while the villains have to Google things all the time, and in Tom Cruise’s words “spend too much time on the internet.” The distinction between usefulness and convenience is a powerful mental model and helps articulate why some AI use feels “wrong” (like atrophy) while other use feels “right” (like an exoskeleton).

• Why Is So Much of Linguistics Math? (Language Jones) — A professional linguist explains how practitioners in his field use fast Fourier transforms, multivariate regression, automata theory, linear algebra, differential equations, constraint satisfaction, and more. Heavy math use isn’t limited to NLP; historical linguistics, phonology, and the author’s own subfield of sociophonetics, to name a few subfields, all require advanced math knowledge.

📚🛋️ Book for your shelf

A book that will help you dip your toes into systems thinking or explore its broader applications.

This week, we recommend The Liberating Structures Fieldbook: Flipping the Script on Meetings, Planning, and Progress by Keith McCandless and Nancy White (2026, 312 pages).

Liberating Structures builds on a simple idea: all of the framework and philosophy in the world won’t go very far without a structure of day-to-day interactions to build it on. But when it comes to large group interactions, the “big 5” approaches of interaction are limiting: the presentation, the managed discussion, the status report,the open discussion, and the brainstorm all tend to favor limited participation and they often focus more on the status quo than on tapping the collective intelligence and motivation of the larger group.

The Surprising Power of LIberating Structures presents 33 different alternate patterns for pulling on broader sensemaking. This ranges from structures like “1-2-4-All” which encourages a tiered discussion to get more participation to using improv inspired techniques to find solutions to problems to using liberating structures to design sessions that themselves utilize liberating structures (how meta!). The first parts of the book rather sells past the close in terms of motivating the structures, but you can easily skim it and jump straight to the patterns.

One note: the book we actually recommend (despite having it on order and not having read it yet!), is The Liberating Structures Fieldbook, the successor to the above two. We are provisionally recommending the fieldbook because it adds 10 new patterns and, critically, updates the patterns for online interactions.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 247 — August 13th, 2026 — Available at read.fluxcollective.org/p/247

Contributors to this issue: Neel Mehta, Boris Smus, MK

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Erika Rice Scherpelz, Jasen Robillard, Jon Lebensold, Justin Quimby, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“In general, we’re least aware of what our minds do best.”

— Marvin Minsky, The Society of Mind (1986)

💎⚾ Crystallized insights from the baseball diamond

It’s an old adage that’s drilled into the head of every young baseball player: “Never make the first or third out at third base.” That is, if you’re standing on second base and deciding whether to steal third – or if you just hit a double and are thinking about legging it out to third base – only do so if there’s one out in the inning (and not zero or two). It’s an oddly specific strategic rule of thumb, but its accuracy is generally borne out by the data.

Though the rule is easy to remember and easy to follow, the trouble comes when a kid asks their coach why it’s true. Most likely, after some fumbling around with rules and partial explanations, the exasperated coach will just have to ask the kid to trust them.

The actual explanation is interesting in its own systems-y way; we’ll mention it at the end of this episode.^[1]1 The upshot is that it takes several minutes to explain and even more time to digest; requires balancing risk and reward probabilities at various game states; and demands an intuitive feeling for the game that your average six-year-old just won’t have. And when you’re hustling hard toward second base, helmet flying off, you typically don’t have the luxury of time to piece together the optimal strategy.

As a result, it’s much easier for everyone to just memorize the one-line catchphrase. This is the advantage of mental shortcuts, or heuristics: they let you compress a hard-to-calculate insight into a clear and easily applied rule. We use these shortcuts every day, from matching our belts to our shoes to not dishwashing our cast-iron pans. We could rediscover everything from first principles or sift through piles of data, but we typically find that the shortcut is good enough and our time is better spent elsewhere.

Of course, the downside of just knowing the rule – without the full logic chain that produced it – is that your reasoning is brittle and not able to adapt to altered situations. Normally you shouldn’t steal third with two outs, but what if there’s a runner on first as well? What if you’d be the tying run? What if the current batter hits a lot of grounders? Armed with our adage alone, you’d have no idea what to do. You couldn’t possibly memorize a special version of the rule for every conceivable situation; at a certain point the scale tips and it’s more efficient to know the game well enough that you can reason through things from the ground up. This is the advantage of intuition and experience: they help you handle novel situations that no heuristic could possibly cover.

The other drawback of relying on shortcuts is that sometimes the underlying logic or facts shift and the shortcut gets “stale”... or sometimes we realize that the conventional wisdom has always been wrong. With modern dishwashers, it’s actually okay (or perhaps better) to not pre-wash your dishes. The old advice about booking flights on Tuesdays no longer applies thanks to airlines’ evolving pricing algorithms. And baseball diehards know that lineup construction has changed due to analytics and new rules: you now want to put your best batter second instead of third or cleanup.

And yet, despite all the flaws of mental shortcuts, we keep turning to them. These crystallized nuggets of advice provide a clear, memorable punch of knowledge that’s still close enough to the complex truth. Perhaps more importantly, they’re easy to transmit – so we end up picking up a lot of them from our parents, teachers, and coaches. That’s the beauty of (lossy) compression, just as it’s easier to email a ZIP file than a giant document or transport a boxed-up air mattress than the unwieldy, inflated version.

Still, we have to revisit these rules periodically and know when to build out our own mental machinery instead of relying on received guidance. Sometimes, it’s helpful to be like that persistent kid who’s always asking their coach “why”: you’ll come away with a more flexible strategic mind, and you might just overturn the conventional wisdom while you’re at it.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🎓 Physics PhDs are leaving the US at record rates

New data shows that nearly 30% of graduates of American physics PhD programs left the country in 2025, the highest rate in 30 years; what’s more, 12% of US citizen graduates departed, a significant increase from 9% the previous year and 5% the year before that. Similar recent studies on brain drain have found that physics and astronomy programs in the US have seen sharp declines in enrollment, and that young biomedical researchers have been leaving the US at increased rates as well. Policy changes are a likely factor – such as restrictions on student visa length, deep cuts to science funding, and rumored $100k fees that may be coming to the OPT visa that many PhD grads use.

🚏🩶 Cars and houses are becoming increasingly gray and beige

American cars are becoming increasingly monochrome: the share of new cars in grayscale colors rose from 60% in 2004 to 80% in 2023, as production of purple cars dropped by 93% and gold by 97%. Observers have noted similar patterns in houses from San Francisco to DC, where gray, white, black, or beige façades are taking over neighborhoods formerly dominated by colorful Victorians. (Small wonder, then, that the top paint colors from Sherwin-Williams include “Crushed Ice,” “Agreeable Gray,” and “Universal Khaki” – the last being its 2026 “color of the year.”) Designers think one psychological reason for the shift to neutrals is “cocooning,” where customers seek safety in purchases: they’d rather have a bland color than a bright one they might regret later. And in a world with ultra-expensive cars and houses, people may also prefer inoffensive colors to avoid risking their items’ resale values.

🚏🌋 Energy IPOs are booming amid the AI datacenter buildout

Semiconductor and memory stocks have cooled down in recent months, so the “picks and shovels” strategy of investing in AI infrastructure has shifted to companies providing power to datacenters, such as those building geothermal or nuclear plants or those making transformers and other electrical distribution equipment. The first half of 2026 saw energy firms raise a record $12.6 billion in IPOs. However, many of these now-public companies are still building experimental technology that’s not commercially viable, and they may be overvalued, as many new energy stocks have seen swift valuation drops shortly after going public.

🚏🧀 An Italian cheese wheel bank is threatened by climate change

One bank in northern Italy stores nearly half a million wheels of Parmigiano Reggiano cheese (worth over 300 million euros) in a giant vault. Dairy farmers give this cheese as collateral for loans they take out; they often need cash advances because the cheese wheels take years to age. But the famous cheese bank is under threat as heat waves across Europe are forcing it to spend ever more on cooling systems and insulation, not to mention that hotter temperatures make cows eat less and thus produce less milk.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Fighting Fire With Fire: Scalable Oral Exams With an ElevenLabs Voice AI Agent (Panos Ipeirotis) — To combat the ease of cheating in any take-home exam, an NYU prof deployed AI voice agents to conduct oral exams at scale utilizing a “Council of LLMs” to triangulate grades. While students found the format stressful and preferred written tests, they conceded it was a more accurate measure of understanding. But will this format accidentally select for oral charisma rather than deep knowledge?

• BRICS in 2025 (Phenomenal World) — Argues that you can get a glimpse of the post-American future by looking at BRICS, the loosely aligned grouping of developing economies consisting of Russia, China, India, Brazil, South Africa, Indonesia, and Egypt. They’re reorienting to an economy built around green energy, electrification, high-tech manufacturing (from EVs to robots to AI), dedollarization, and diverse trade partnerships. As the US moves back toward fossil fuels and protectionism, you can increasingly see the two worlds pulling apart.

• Is Switzerland Tired of Prosperity? (The Guardian) — De Weck argues Switzerland’s looming referendum to cap population/immigration is self-sabotage: openness, not control, built its prosperity. He invokes Thomas Mann’s Buddenbrooks framing that prosperity breeds complacency, eroding the pragmatism that built it, and notes that Nestlé, Swatch, and Novartis, the most “Swiss” of institutions, were themselves built by immigrants.

• Preliminary Theory of the In-Group Contrarian (Outsider Theory) — Applies philosopher René Girard’s mimetic theory (where we desire what we see our group members desiring) to social media, exploring how many platforms see the rise of an “In-Group Contrarian,” who shares the values of their broader group but objects to certain actions. The group may love to pile on the contrarian, but they can rarely expel them, because “hatred of the [contrarian] can become one of the main things that binds the in-group together.”

🧩🕹️ Systemic tools and toys

Games, gadgets, demos, and other nifty interactives to stretch your systems-thinking capacity.

By 2080, the climate in New York City will feel like that of modern-day central Arkansas; Bay Area residents will feel like they’ve moved to deep SoCal. That’s what a fun interactive map from the University of Maryland Center for Environmental Science shows: pick anywhere in the world and you’ll see what current-day location matches its projected 2080 climate, assuming we don’t get emissions under control.

It’s interesting to click around and see which areas will be affected the most or the least. For instance, you’ll see huge “movements” in eastern North America and central Europe, but hardly any around coastal South America or South Africa. Try to find surprising shifts – we won’t spoil the result, but try clicking on Miami, Shanghai, or the Yucatán Peninsula. For an extra shock, try clicking anywhere in the Amazon basin, Singapore, or the Persian Gulf.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 246 — August 6th, 2026 — Available at read.fluxcollective.org/p/246

Contributors to this issue: Dart Lindsley, Ade Oshineye, Jasen Robillard, Boris Smus, Justin Quimby, Erika Rice Scherpelz, Neel Mehta, MK

Additional insights from: Ben Mathes, Jon Lebensold, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“We know they lie, they know we know they lie, they don’t care. We say we care, but we do nothing. And nothing ever changes. It’s normal.”

— Adam Curtis, Hypernormalization

👉👈 I know you know I know

Look at that chart. The numbers are baaaaaddd. We all know it. But no one says so. It’s not ignorance. It’s a failure of our shared reality. We all know the project is in trouble... but until someone (or, perhaps, the right someone) says it aloud, the status quo is frozen.

If I know something, it’s individual knowledge. If everyone knows, it’s mutual knowledge. If we all know we know, it’s common knowledge. These layers of knowing seem academic, but our social lives depend on our deep and often intuitive knowledge of them. (Here’s a useful video on these topics.)

When something becomes common knowledge, such as through a public announcement, it changes what people can safely assume. It can create a Schelling point, where we not only can make a decision, but can generally expect people to reliably do the same thing, such as exiting a building when an alarm goes off.

Mutual knowledge can still be valuable. If we all know the goal we’re pursuing, then we can row in the same direction. Yet it is also the realm of kayfabe, where we all act as if something is true even though we know it’s not. That’s our bad chart example. Kayfabe isn’t exactly a lie. It’s more a coordinated agreement about which truths are allowed to have power.

When the shared fiction becomes common knowledge, its nature can change. Sometimes that’s a positive. Imaginative play is fun because we’re all buying into the “lie” that the couch is a submarine or the floor is lava. In organizational settings, though, it can drift toward an open secret. Everyone knows that (nearly) everyone knows... but they lack permission or power to change it.

Systems built on open secrets can look stable... but only because everyone is avoiding conflict to keep the machinery moving. At small scales, this can be functional. We all know that some processes are low value but not worth the effort to change. The trouble begins when the shared fiction prevents the group from updating its knowledge or actions as circumstances change. People begin optimizing for what wins them success in the game rather than what creates success in reality.

This can become costly, even more than a falsehood. A falsehood can be corrected. An open secret recruits everyone to its maintenance. And the more people who are involved in the maintenance, the harder it becomes to be the person who speaks up.

When coordinating people, it’s not enough to think about what you know or even what we know. We need to understand what’s common knowledge—and notice when that shared understanding gives us room to act. Which facts in your world are merely known? Once something becomes sayable, the group gains more choices about what to do next.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏😈 Japan made “cruelly hot day” an official weather category

The Japan Meteorological Agency uses memorable terms for warm days to encourage residents to protect themselves: days above 25° C are “summer days” (natsubi), 30° C days are “midsummer days” (manatsubi), and 35° C days are “extremely hot days” (mōshobi). But with temperatures above 40° C (104° F) becoming common in Japan—the country topped 40° C on nine days last year – the JMA has introduced a new word for days above 40° C: “cruelly hot day” (kokushobi). The term is already coming in handy as Japan suffers under another heat wave: the country reported five straight kokushobi in July, and nearly 11,000 people were hospitalized due to heat-related illnesses in a single week.

🚏📝 An AI-proctored entrance exam had so much cheating that 58k students have to retake it

This May and June, nearly 160,000 students took the entrance exam for UNAM, Mexico’s largest and arguably most prestigious university. This year’s exam was administered remotely (the first time ever) and required students to install “lockdown” software on their computers that would prevent them from visiting other sites or apps along with an AI system that monitored test-takers’ faces via their webcams. Despite these safeguards, top scores were far more common than usual: 16.3% of students scored 100+ on the 120-question test (vs. 3.5% in 2021-2025), and 5.5% got 110+ (vs. 0.9% in 2021-2025). The rise in high scores led to widespread allegations of cheating; AI researcher Raúl Rojas suggested that “almost half of the students” were cheating. So, UNAM is requiring everyone who “passed” the original test – a full 58,000 students – to retake the exam in person.

🚏🌀 We may experience the strongest El Niño ever recorded

The El Niño–Southern Oscillation (ENSO) is a recurring meteorological event driven by complex feedback loops originating from weaker trade winds across the Pacific that leads to varying patterns of droughts, flooding, and extreme temperatures in different parts of the globe. The strength of an El Niño event, which tends to happen every few years, is often measured by Pacific sea surface temperature (SST) anomalies. The strongest ENSOs ever recorded hit SST anomalies of +2.75° C, but weather models predict that this year’s event (which is just heating up) will peak at +3.6° C, blowing past the record; one meteorologist wrote that it could “perhaps be the strongest in 500 to 1,000 years.”

🚏🐷 LinkedIn added a “seems like AI slop” button

A few years ago, LinkedIn introduced buttons to let users draft or edit posts with AI. The AI slop epidemic has apparently gone bad enough, though, because the social network is dumping that feature and instead rolling out a new button that lets users report posts as potential AI slop. LinkedIn also says it’s introducing classifiers (trained, in part, on these button presses) to automatically identify and downrank potential AI-generated content.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Damming the Big Ocean (NY Review of Books) [Archived] — A review of a political economy book on chokepoints’ role in economic warfare. It argues that the US’s “economic arsenal” of tariffs, sanctions, embargoes, export controls, etc., originally used to combat the USSR and later terrorists, has been refashioned into crude weapons against economic rivals like China. The problem is that the US has lost its leadership in free trade and globalization as erstwhile allies have turned elsewhere, “leaving [the US] more isolated and in control of a shrinking share of the global economy—reliant on undependable policy tools to preserve its edge.”

• Beware Software Brain (The Verge) — Coins the term “software brain”: the tech-industry delusion that messy human reality can be reduced to structured data and executable rules, and that AI can then automate it. Editor-in-chief Nilay Patel argues the AI backlash isn’t a marketing problem, but rather downstream of the lived experience of being flattened into a record in someone else’s database.

• ENSO Essentials (Columbia Climate School) — Examines the complex climatic systems that give rise to the El Niño–Southern Oscillation and its inverse, La Niña: the ‘default’ state of the Pacific Ocean includes a positive feedback loop between the ocean and atmosphere (a “coupled system”), while El Niño and La Niña are also positive feedback loops, but they temporarily disrupt the original loop. So the system can be thought of as transitioning among three different positive feedback states.

• The Inner Robot (Matt Mason) — By examining the human ulnar grasp, one of many complex manipulations our hands are capable of, the author observes that we seem to have two distinct modes of operation. The first is “in our heads,” where we are operating deliberately and in complete control, good for planning and learning but weak at performing tasks. The second is our “inner robot,” an autopilot run by the subconscious that excels at repetitive, often physical tasks like object manipulation.

🔍📆 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: repugnant transactions.

A shop down the street sells medication 24/7. A customer arrives late at night to buy medicine for their sick child. The catch: the shopkeeper is selling prescription medications without requiring a prescription. On one level, this sale is compassionate and practical.

But it’s also the sort of sale that authorities restrict. And for good reason: The medicine could be counterfeit. The buyer could misuse it (intentionally or through misdiagnosis). The restrictions protect in aggregate even if many particular cases can be justified.

Repugnant transactions are transactions wanted by the parties involved but restricted by a third party. There are many reasons for such restrictions, including mitigating externalities, enforcing moral norms, and preserving political power. Whatever the reason, those directly involved don’t have the final say. Such transactions will often move underground, giving rise to a repugnant market.

There is a mirror version, where two parties may dislike an activity but a third party forces it upon them. We might call these conscripted transactions. A doctor and a patient may both think an annual exam to renew a medication the patient has been safely using for years a waste of time, but if insurance requires it, they’ll do it. It’s the same pattern of third-party involvement, in the opposite direction.

If your goal is to disrupt a space that seems oddly ripe for it, understanding what transactions might be repugnant or conscripted will help highlight if the problem is a lack of product-market fit… or if it’s the shape of the market. When the market seems to be underinvesting or overinvesting in certain types of transactions, look for a third party to see if there might be an external reason why.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 245 — July 30th, 2026 — Available at read.fluxcollective.org/p/245

Contributors to this issue: Justin Quimby, Erika Rice Scherpelz, Neel Mehta, Boris Smus, MK

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“For our house is our corner of the world.”

— Gaston Bachelard, The Poetics of Space (1958)

💸📍 No cheaper way out

It’s time for a nice relaxing weekend at your vacation home. You settle into your remote sanctuary, watch the sunset over the water, feeling grateful for such an immense privilege... and then reflect on how an AirBnb would cost a fraction of the vacation home’s mortgage and upkeep.

But that’s not how it actually goes.

While a vacation home can become a burden, we think of owning and renting differently. The house you own is a home with memories. It is what economic sociologist Lucien Karpik calls a singularity. Not in the black-hole sense. And not in the AI sense. A singularity, in his sense, is a good whose value is tied to its particular, hard-to-reproduce characteristics. A rental may be cheaper, but it’s not the porch where your son took his first steps. It doesn’t have that mark on the wall where your daughter crashed her tricycle.

Singularities are one of the many sources behind price inelasticity. Price elasticity tells you how much demand responds to price, but not why some goods resist that pressure... or what might cause that to change.

For singularities, price inelasticity is fairly substitution-proof. The existence of AirBnbs may reduce the demand for vacation homes generally, but it doesn’t affect personal demand for the place your grandparents built before you were born. Luxury goods can also generate inelasticity. For these and other status items, inaccessibility (including price) is part of the point.

Other scenarios aren’t so sticky, and demand can change quickly if a viable alternative appears. A necessity like healthcare or education is subject to price inelasticity because the human capacity that provides it is limited. On the other hand, few people kept horses once cars and tractors became available. Other less-sticky sources of price inelasticity are system constraints (like insurance networks, accreditation rules, licensing), switching costs, and urgency.

Sometimes sticky and non-sticky cases overlap. “A doctor” may be a necessity, but “my doctor” adds history and trust. And we sure want our doctors to have those credentials.

This variety of factors is why Baumol’s cost disease lands as a crisis when it comes to healthcare or education, and as the price of luxury when it comes to fine dining or a live show. In the first category, prices are stuck but we would switch given a viable alternative. In the second, our willingness to pay is driven by what we find personally meaningful.

These two paths look similar on the balance sheet, but they feel very different when you’re paying the bill.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🧯 More than half of England is in drought conditions amid “climate whiplash”

The UK is currently struggling under its fourth heatwave of the year; July 2026 has been the country’s second-warmest month on record. Relatedly, most of central, eastern, and southern England, as well as northern Wales and eastern Scotland, are in a state of drought or extreme water scarcity. This comes just a few months after a very wet British winter where much of England saw twice as much rainfall as usual in February — this “climate whiplash” has led to dozens of wildfires across the country, as a wet winter drives the growth of vegetation that becomes a tinderbox in a dry summer. Scientists note that this pattern is consistent with a warming atmosphere, which can hold more moisture: it’s like a big “sponge” that absorbs a lot of water from the ground (leading to droughts) but then releases large amounts at once (leading to floods).

🚏🎰 Americans lose more to sportsbooks than they spend at the movies

Last year, Americans placed $166 billion in bets on sporting events, which is more than double what they spent on movie tickets ($8.9 billion), recorded music ($11.5 billion, a record figure), live music ($18.5 billion), books ($14.6 billion), and museums ($16.4 billion) combined. Economists believe the $166 billion figure may still be an undercount, as it excludes betting at tribal casinos (which could tack on another $5 to $10 billion) and on prediction markets (which could add $50 to $100 billion). Added up, it means that the average American adult bets almost $1000 a year on sports, though the market is dominated by big spenders, with 5% of bettors taking 95% of the losses.

🚏🔋 Indian car companies are making more energy-efficient EVs than BYD or Tesla

Chinese EV maker BYD is frequently in the news for its rapid growth and cheap electric cars, but Indian automakers have been doing quite well under the radar. Two of the largest Indian car manufacturers, Tata Motors and Mahindra, topped a recent leaderboard for EV energy efficiency, measured as the number of watt-hours needed to travel one kilometer across a company’s entire battery-electric fleet. They beat out Tesla (3rd), BYD (4th), and a slew of legacy manufacturers. Electric cars account for under 5% of new car sales in India, a far cry from the global average of 25%, but the country is reportedly working on new emission standards that may boost EV sales in the 2030s.

🚏🎙️ Arizona State is launching a bachelor’s degree in being an influencer

ASU’s journalism school has started a new BA in Content Creation, which will prepare students to become “an influencer and strategic storyteller.” The curriculum includes courses on videography, editing, podcasting, on-camera presence, personal branding, audience analytics, and social media strategy, and the program’s capstone project will require students to create a social media account and grow their follower count.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• How Funerals Keep Africa Poor (David Oks) — Uses Africa’s ruinously lavish funerals, which can cost more than a year’s salary, as a lens into the political economy of kinship networks. These tribal groups are a sort of “mutual aid society that you’re born into and can’t leave”: productive and high-earning members are expected to financially support the less productive. If anyone gets too wealthy, they may leave the network, which threatens everyone else, so these societies have developed intentional wealth-destroying traditions like fancy funerals or the Pacific Northwest’s famous potlatches to ensure that nobody accumulates too much surplus. However, this stifles ambition and innovation and makes economic growth all but impossible.

• Engineering Management After the Cost of Code Collapsed (Karim Jedda) — Argues that many of our foundational assumptions about business hierarchy, KPIs, and management need to be revisited now that code is nearly free to produce. A general pattern is that “when generation is abundant, verification is the constraint.” On the engineering side, it means that we can get great benefits from investing in machine-checkable correctness such as automated tests; for management, our old roles as information routers and presentation makers are obsolete and the job boils down to spending all day on decision-making and accountability, because that’s the part that LLMs can never do.

• Intellectuals Are F*cking Idiots (Mark Manson) — Argues that smart people keep failing in public because they’re rewarded for elegant models, not for matching reality, and they forget the map isn’t the territory. Blue Zones with people living to 100+ are due to dodgy birth records, Ehrlich’s population doom never came, and Pol Pot killed the professor who lectured him. The real dividing line isn’t intelligence but whether you update your views when the evidence changes and hold your opinions lightly.

• Supermarket Economics (Phenomenal World) — A deep-dive into the financials of supermarkets and grocery stores, including the history of and rationale for mega-mergers; the K-shaped bifurcation in the industry (with Whole Foods and Sprouts at the high end and Walmart and dollar stores at the bottom end); why grocery delivery and e-commerce haven’t panned out yet; and how the countercyclical nature of grocery stores makes them popular investments with very good return on equity.

🔮📬 Postcard from the future: Progress from Nihilism

A ‘what if’ piece of speculative fiction about a possible future that could result from the systemic forces changing our world.

// What might be the consequences of AI-accelerated engineers becoming the norm?

// September 2107

// A group of teenagers gather around a wizened lady in her rocking chair

“Grandma, can you tell us a story about the ‘American Troubled Times’?”

“Ah, yes. You’ve gotten to that part of your history class! I always wondered if you were going to ask.”

She gestures to the sprawling complex, housing several multi-generational families amidst vegetable gardens, solar arrays, local compute nodes, utility relays, and transit docking stations while a cargo airship drifts overhead.

“This space and environment is the result of many hard fought battles in the face of massive waves of change and trauma. Nod if you’ve heard these terms in your lessons: Warsaw Riots of ‘37, the Helium crunch of ‘42, Smokey Mountain Inferno of ‘29, Excel riots, Looting of Project Rainier, the MSCI bankruptcy, the Panama Canal dry out, Bankruptcy of Social Security? Each of these would have been considered a major world event in the second half of the twentieth century, but they all happened within 15 years of each other.

“So what were we, as a generation, supposed to think and believe in the face of such massive change? The short answer is that for a while, many of us embraced nihilism. Not in the form of Moral Nihilism where nothing is objectively right or wrong, and that moral truths do not exist, but instead a nihilism about the truths and beliefs of the 20th century.

“Many ‘Gods’ died in that period. Beliefs, Governments, Schools, Companies. And in some cases, the people who led those things died as well. Many of those who held to those belief structures too tightly were unable to surf the waves of chaos and uncertainty in the years that followed and drowned.

“Those of us who embraced the chaos and shed ourselves of attachments to the old ways were able to survive, and in time, thrive. We built new structures of governance and finance, trying to not repeat the mistakes of the past, but knowing that we would make new mistakes. Which led to this…

“I hope you can continue that change.”

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 244 — July 23rd, 2026 — Available at read.fluxcollective.org/p/244

Contributors to this issue: Neel Mehta, Erika Rice Scherpelz, MK, Boris Smus

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Justin Quimby, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“Poets are not to blame for how things are.”

— Emily Wilson, The Odyssey

🧑‍✈️🔗 How trouble travels

You’re flying from NYC to Cabo, connecting in Dallas. The flight gets delayed four hours because of mechanical issues. That’s okay, you’re still going to Cabo. Wait, the gate just changed and there’s a new plane. Fortunately, you make it to the gate on time and you finally board.

Hurrah!

Except… not. By the time you’ve boarded the plane, the pilots have surpassed their duty time limitations. Now you have to wait for a new crew to come, leading to even more delays. They finally arrive and the plane finally gets going… just moments before a 15-minute thunderstorm throws the takeoff schedule into chaos for the next three hours. (And good luck if you missed your connection in Dallas by even five minutes, which may force you to wait several hours for the next flight to Cabo.)

Aviation runs on such tight margins that the schedule has no room for error. This leads to the system becoming very brittle. In general, tightly allocated resources, be they time, supplies, or anything else, can lead to cascading failures. Cascading failures aren’t simply when one failure comes after another, but specifically when the interconnectivity between parts of a system causes one failure to trigger the next.

The more tightly a system is put together, the more prone it is to cascading failures. A physical manifestation is dominoes. Put them close enough, and the “failure” to stand up propagates, often to impressive ends. Put them sufficiently far apart, though, and the isolation prevents individual failures from propagating through the whole system.

As the domino example shows, the mechanics of a cascade need not be negative. The generalization, where a tightly coupled system makes it easier for state to propagate through the system, can be used to positive effect if you want to spread something. The sociological concept of a tipping point is broader and is often used in positive contexts, such as the increasing popularity of a product.

But in the sadly frequent case where you truly are talking about cascading failures, what can we do? A common technique is to introduce slack: extra time, extra resources, or extra capacity that can be used to cover the gap when something goes wrong — and something always goes wrong. Too much slack is inefficient, but too little slack can be even worse.

The broader principle is that coupling can be a source of value or a source of risk. By understanding how much our system is coupled and where we have opportunities to increase or decrease that coupling, we can adjust — though not control — our resilience to cascading state changes.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🗳️ More Congresspeople are leaving their seats than in any year since 2000

So far, 11 members of the US Senate and 59 members of the House of Representatives have not filed to run in this November’s federal elections. This total of 70 members not running is by far the most since 2000. More Republicans than Democrats are stepping down: 7R-4D in the Senate and 36R-23D in the House. Note that many Representatives are aiming for higher office (Senate and governor), but a full 27 Representatives aren’t running for anything else — the most retirements since 2012.

🚏🛢️ Tech giants are moving data out of the Gulf along oil pipelines

Most of the internet cables connecting the Gulf states to the rest of the world run underwater through the Persian Gulf, Strait of Hormuz, and Red Sea. The war in the Middle East has put these cables at risk, so cloud giants like Google and Amazon have been turning to a backup: fiber-optic cables that take a land-based route from Iraq to Turkey and onward to Europe. These cables run alongside oil and gas pipelines, so they can piggyback on the existing right-of-way and security infrastructure. (However, even if the fiber routes are taken care of, data centers in the region remain at risk, as Iran has already bombed AWS data centers and caused outages.)

🚏☕ Seattle’s job market is shifting from tech back to frontline/service work

Washington State suffered the second-most tech layoffs (11,000+) of any state from May 2025 to April 2026, only trailing California. A new report looked at job postings in the Seattle area and found that the Emerald City has shifted sharply away from its ‘traditional’ driver of big tech jobs and toward frontline and service workers such as truck drivers (+ 231% in the last 30 days), security guards (+146%), and baristas (+30%). Meanwhile, openings for software engineers fell 56% and those for product managers were down 19%. The growth still outweighs the declines, as the region’s overall unemployment rate dropped from 5.7% to 5.0%. However, economists warn that some of the service-sector hiring boom may have been transient due to the soccer World Cup.

🚏⚡️ Autonomous micro-drones can shoot down mosquitoes from the air

Large combat drones have made waves in the Ukraine war, but micro-drones have been finding uses in an unusual domain: pest control. One startup has created a new gadget that uses a “phased array sonar base station” with hundreds of microphones to map a small area down to tiny scales and identify mosquitoes by their “unique wingbeat signature.” Then, a computer sends remote commands to a 40-gram aerial drone that kills identified mosquitoes up to 8 meters (26 ft) away.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Underrated Ways to Change the World (Adam Mastroianni) — Argues that you don’t need the soul-crushing wealth of “earning to give” or the saintly sacrifice of scrubbing orphanage toilets to improve society. Meaningful progress often comes from overlooked, low-status middle paths. Mastroianni champions “first followers” who validate whistleblowers to expose massive frauds, social “switchboarders” who forge a scenius among isolated idealists, and everyday amateur researchers who use their total freedom to solve overlooked problems. The piece is an inspiring defense of the unglamorous maintainers — rather than the founders or billionaires — who actually keep civilization running.

• Who’s Afraid of Frontier Models? (Stratechery) — Ben Thompson argues that the cost structure of serving AI models, where marginal costs for inference are high, requires us to analyze the LLM industry more like conventional commodity manufacturing than like zero-marginal-cost tech businesses. The upshot is that American frontier LLM labs shouldn’t be too worried about Chinese open-weight models undercutting them, because it’s unlikely that the latter will be able to achieve a substantial edge on marginal cost. Still, he argues that the US should spur the development of homegrown open-weight models by removing restrictions on distilling (which is a big part of how they can be trained so cheaply).

• The Vibecession Hasn’t Gone Away (Matt Darling) — Shows data suggesting that the gap between actual economic conditions and self-reported consumer sentiment (the “vibecession”) has only increased since 2021, rather than being a temporary COVID trend. The gap is visible across all income classes, and in fact it’s slightly higher among wealthier Americans. The author draws on Polish economist Michał Kalecki to theorize that the dissatisfaction we’re seeing comes from how low unemployment drove up wages among lower earners, which increased costs for services like health care and food delivery. This may tell an uncomfortable story about the political ramifications of workers gaining bargaining power.

• Movin’ On Up: How Costly New Homes Create Affordable Old Homes (California YIMBY) — Examines two mechanisms whereby building new market-rate (or even luxury) homes ends up increasing the supply of affordable homes. “Filtering” is a slower method whereby new houses gradually depreciate and become available for lower-income buyers. A quicker mechanism is “moving chains,” where wealthier people immediately move into pricey new homes, freeing up their old homes for the less wealthy, and so on down the food chain.

🔍📆 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: terminus post quem and terminus ante quem.

Suppose you’re rummaging through a desk drawer and find a nice photo of you and your friends, but you don’t remember when it was taken. Fortunately, you can use context from the photo itself and its surroundings to help you date it.

For instance, if your friend is holding an iPhone 15 in the photo, you know it can’t have been taken any earlier than the iPhone 15’s release date — this gives you what’s known in archaeology as the terminus post quem, which is Latin for “time after which.” That is, it’s the ‘early bound’ on when the photo was taken.

Meanwhile, if you found the photo buried underneath tickets to a baseball game from a certain date, you can assume it was taken on or before that day — this is the terminus ante quem, or “time before which.” However, this dating is less secure: it relies on the assumption that the relative ordering of objects reflects the order in which they were placed. In other words, you assume that nobody has shuffled things around and that you didn’t (for some reason) bury the photo deep in the drawer when you initially dropped it there.

Our photo example was a basic one, but this pair of concepts is extremely important in archaeology, where you’re constantly faced with layers of buried artifacts that don’t advertise their origins and where even a few decades’ swing in the date of an artifact can completely change the narrative you’re constructing. And you’d be surprised how often these concepts can help you sleuth in your daily life: when was this undated corporate strategy doc written (and thus, did it account for market shift X or reorg Y)? What year did Grandpa graduate from college again? Or even, how long have these leftovers been sitting in the fridge?

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 243 — July 16th, 2026 — Available at read.fluxcollective.org/p/243

Contributors to this issue: Neel Mehta, Boris Smus, Ade Oshineye, Erika Rice Scherpelz, MK

Additional insights from: Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Justin Quimby, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“It is a profoundly erroneous truism… that we should cultivate the habit of thinking of what we are doing. The precise opposite is the case. Civilization advances by extending the number of important operations which we can perform without thinking about them. Operations of thought are like cavalry charges in a battle — they are strictly limited in number, they require fresh horses, and must only be made at decisive moments.”

— Alfred North Whitehead

🥗🎁 The beauty of paying it forward

When invited to a potluck, how does everyone figure out what to bring? For example, you might bring your specialty homemade lasagna while someone else shows up with a bag salad and another person provides the wine or a stack of plates. The host obviously provides the space, and if everyone is lucky, that much rarer gift of actually remembering everyone’s name (and introducing everyone to each other).

It’s clear that everything at that table has a value, in cost, attention, and effort, and it’s plain to everyone seated at said table. However, it would be pretty strange if someone went through and priced every part of it. Still, *something* is in effect because the unfortunate freeloader who arrives empty-handed three times running will get noticed (and maybe gently admonished or even conveniently omitted from the next event’s invite list). That said, the system only works because nobody converts all of it into explicit units of exchange and value. For the moment we do, the whole thing falls apart and ends up no longer being a party.

Anthropologists call the above dynamic *a gift economy*, where value circulates through the ordinary machinery of mutual aid: hospitality, reputation, and everyday care. Gifts are able to build trust precisely because they keep moving from one hand to the next without the need for expensive, inefficient coordination and negotiation. Consider how a mentee could repay the mentor who changed their life for the better years ago: to cut them a check would be gauche so instead, they repay their mentor by helping someone else down the road without ever calling it (or thinking of it as) repayment.

Despite this, gifts obviously still create obligations; they just aren’t the clean, legible obligation of debt and value exchange. While a debt demands speedy repayment and closure, a gift wants to live on and on.

That imbalance is the point and the consensual forcing function. It powers an open loop that keeps people coming back to the table, to keep the party going.

Of course, it’s not all conviviality and revelry: the open loop of paying it forward can cast long shadows. Over time, generosity can curdle into unspoken expectations that fall unevenly (usually on whoever, due to either social pressures or individual differences, finds it hardest to say no). One benefit of accounting is to prevent exactly this: a ledger makes obligations legible and legible obligations can be fairly shared.

As a result, we might be tempted to run the potluck like a balance sheet. And for a while, this balance sheet party might even work. But when everything is reduced to balances and debts, the community loses the very thing it was for. Accounting makes a system more legible, but also less alive.

The trick isn’t choosing between the ledger and the revel. It’s understanding each one’s comparative advantages and tradeoffs. Ledgers close loops while communities keep them open: knowing that, what might we be tallying right now that would circulate more freely as a gift?

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🥬 The US stopped monitoring the “explosive diarrhea” bug last year

Cyclosporiasis, better known as explosive diarrhea, has been sweeping across the USA, with over 7,000 cases reported so far across nearly every state. Sharp-eyed observers noticed that the US’s Centers for Disease Control used to monitor cyclospora, the parasite that causes the illness, but stopped monitoring it last year, along with listeria, shigella, and other bugs.

🚏🇨🇺 Solar-powered tricycles are taking over Cuba amidst a fuel crisis

Cuba has been virtually unable to get oil or gasoline since the US threatened tariffs on any country that sent it oil; just a single oil tanker has arrived in Cuba since January. Public transportation has been shut down and residents are suffering under prolonged blackouts. An unlikely savior has emerged: cheap electric tricycles from China, which blanket the island country’s streets and are being used for everything from package deliveries to taxi services to trash pickup. (Indeed, they’ve largely supplanted Cuba’s iconic vintage cars.) Some enterprising drivers are even installing solar panels on awnings atop their tricycles to recharge on the go.

🚏🧓 The US is facing a looming workforce shortage as Boomers retire

From 2024 to 2032 (when the last Baby Boomers will likely have retired), 18 million Americans are projected to leave the workforce, while fewer than 14 million will join — which one firm is calling “the largest labor shortage the country has ever seen.” Major contributors include the aforementioned retirements, declining immigration and birth rates, and a mismatch in what college grads are studying — for instance, the country is producing too few healthcare grads and too many business/finance grads. The shortage is uneven across states: 40 states currently have fewer workers than open jobs (South Dakota has the fewest, with 41 workers for every 100 openings), and just 10 have more workers than open jobs (California has 153 workers for every 100 openings).

🚏🧑‍🏭 Factories are offering flexible schedules to overcome worker shortages

In the depths of COVID, the American manufacturer GE Appliances was struggling to get workers to come into its plant in Georgia, so they turned to a temp staffing firm that could plug in flexible workers for a few hours at a time. It worked, even though manufacturing’s typical culture is one of a static workforce that puts in 40 hours every single week: the plant recently got a $180 million expansion, adding 600 new jobs. Years later, the plant is still embracing flexible staffing; a pool of 900 workers can pick arbitrary 4-hour shifts via an app. This unconventional “Uber for manufacturing” approach has been a hit among employees and has improved retention: workers who would’ve otherwise retired can now shift to part-time work. Besides GE, manufacturers like Stanley Black & Decker and Georgia-Pacific have also started adopting this model.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Personal Business (Are.na) — McLuhan absolutism strikes again: the funding source determines the product. VC-backed software optimizes for scale and dominance, which structurally produces the extractive, impersonal outcomes we associate with Big Tech. The Are.na cofounder’s advice is to find something that is “genuinely fun and interesting and you know you could be interested in it forever,” set modest scaling goals, focus on community, and bootstrap.

• On Compositionality (Jules Hedges) — An excerpt from the author’s interesting PhD thesis on ‘compositional game theory.’ Here, the author argues that systems with emergent behavior (like those of biology and economics) are the opposite of compositional systems (like computer programs or well-structured bureaucracies), where the whole is exactly equal to the sum of its parts. While emergent behavior is cool (especially to us systems thinkers), it makes a system much harder to reason about, and thus compositionality “is so powerful that it is worth going to extreme lengths to achieve it.” In programming, for instance, we achieve the desired modularity by minimizing the use of global state or “goto” statements.

• Authentic Food? Take It With a Grain of Salt (PCC Community Markets) — Observes that every “authentically Italian” tomato (from the Americas), “emblematically Irish” potato (native to South America), “quintessentially British” fish and chips (fried fish brought by Sephardic Jews fleeing the Spanish Inquisition), and “traditional” taco al pastor (Lebanese shawarma technique, via 19th-century immigrants to Mexico) is an artifact of migration and cultural collision, making authentic cuisine less a heritage to protect and more a myth that collapses on contact with history.

• How Ukraine Built a War Fighting State (Austin Vernon) — Describes how the small, scrappy country has held its own against Russia through a combination of savvy military reorganizations that let the best units and generals rise to the top; heavy investment in drone manufacturing; and exclusive access to unjammable Starlink internet, which lets operators remote-control drones and see real-time video feeds, even in areas where radio links are weak. One interesting organizational technique was building a ‘marketplace’ where units can buy equipment and supplies; the ‘currency’ was “video-confirmed kills” of enemies. This lets the most effective units get more resources and thus get even stronger.

🔍📆 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: stag hunt.

The team agrees that if we build a generic search system, it could replace a whole pile of one-off hacks. It will be a lot of work, but it will make a whole category of backlog items nearly free. Work starts, then slows. We keep adding one-offs: a tag-specific search, an author filter. They only take a week each. The search system is never built.

This is the Stag Hunt dynamic. If they cooperate, two hunters can catch a stag, which feeds them well. Alone, each could bag a hare. It is not a gap between good and bad (like in a Prisoner’s Dilemma). It is the gap between good and better. Cooperation is stable as long as there is sufficient trust that each of us will do our part. If not, the rational decision is to get the hare.

In organizations, failures of cooperation are often stag-hunt-style small, everyday failures of confidence. When we go after hares, we may feel like we are making the best decision, not seeing how short-term optimization weakens longer-term commitment.

This can become an organizational climate, a shared sense that the organization is bad at commitment. Repetition reinforces the underlying pattern. If we consistently contribute to ambitious goals, the organization gets more ambitious. If we consistently hedge, small goals within our personal control become the rational choice.

In bigger systems, such as international relations, this shows up as a lack of ambition or an inability to believe that big things are possible through cooperation. In smaller systems, such as teams without psychological safety, this dynamic shows up as a belief that everything is a Prisoner’s Dilemma. However, successful systems succeed by convincing the participants to focus on growing the pie rather than merely increasing their slice.

One of the reasons we are writing about the Stag Hunt lens is to expose our audience to the rich catalog of empowering tools provided by Game Theory. These tools give us the freedom to reframe an unhelpful payoff matrix into one that empowers us.

For example, the transition from Prisoner’s Dilemma to Stag Hunt shows that by building trust and commitment we can turn a zero sum or negative sum situation into a positive sum situation. We can do this by making commitment visible: naming the larger goal, clarifying what each of us is depending on, and making the cost of ‘playing small’ legible before the temptation of the easy hare appears.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 242 — July 9th, 2026 — Available at read.fluxcollective.org/p/242

Contributors to this issue: Neel Mehta, Boris Smus, Erika Rice Scherpelz

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Jon Lebensold, Justin Quimby, Lisie Lillianfeld, MK, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“True education is a kind of never ending story — a matter of continual beginnings, of habitual fresh starts, of persistent newness.”

— J. R. R. Tolkien

📝 Editor’s note: Our last issue was delayed due to publishing issues. We apologize for the delay!

🐒🦍 When everything is at scale

Note: Today’s piece dives deep on code, but that’s mainly because code is further ahead in this disruption than many other domains. The pattern of “what happens to our understanding when the decisions were made without human supervision?” will apply more and more broadly as more areas become subject to massive AI disruption.

AI is writing more of our code… so what happens when no human holds the whole system in their head? When no one remembers why the retry logic backs off that way, or what — if anything — that defensive ‘if’ still guards. We fear we’ll be maintaining code we no longer understand.

It’s a real concern. It is also, mostly, a eulogy for something we never really had at scale.

Folks who have spent time in a large scale production code base — especially an older one — know that many critical subsystems were already black boxes. The person who made the original decision left long ago. The undocumented constraint already changed. When the debt was built up by humans, we held hope that someone would remember the when and why of it, but that was usually more dream than reality.

So what actually changes now? Our intuitions do have some truth. Managing a large scale code base has changed. AI does not magically abolish the difficulty of maintaining a large code base, nor does it make it uniformly worse. It redistributes the challenge — some of it easier now, some of it worse.

To take another example, self-driving cars are, on average, safer than the humans they replace. But when they fail, they fail strangely such as stopping dead in an empty intersection or mistaking a truck for sky. A human’s bad code smells bad in familiar ways: out-of-date comments, misleading variable names. AI generated code loses those smells. So when things fail, it often feels weird: how can code that looks so good be so wrong?

Some things get better. Documentation, decision logs, and tooling can get better because they get cheaper. Chat transcripts have the potential to capture a lot of the intent that was traditionally lost. But we should remember the limits of these tools: the effort it took before might have been hiding a deeper problem of making sure artifacts provide actual value.

The solution was never to understand everything. As we figure out how to move forward, we will reach for the familiar tools that let us change systems we don’t understand: CI/CD, documentation, automated testing, observability and monitoring, etc.

While it’s true that some of this will require novel responses, we’ll also find that the systems that had the best practices in place before will be the ones best able to absorb the change. We might now need to treat our weekend project like something written by a team of twenty, but fortunately we have their experience to learn from.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🐍 Amazon will stop accepting new customers for Mechanical Turk

Amazon’s venerable Mechanical Turk marketplace, where companies can pay freelance workers a few cents to do manual tasks like data entry or filling out forms, seems to be winding down, as Amazon has announced that it’ll stop accepting new customers for the service. LLMs can now do many of those basic tasks, and ironically, a 2023 report found that more than a third of MTurk workers were using AI to automate the supposedly human work they were doing. (Many ‘turkers’ have moved on to other freelancing sites, so some demand for this kind of work still exists; ‘requesters’ have also apparently moved on due to rampant “bots and fraud.”)

🚏🏗️ China is storing surplus wind power by stacking concrete blocks up to 40 stories

Storing excess wind power for when the breeze isn’t blowing requires some sort of battery — but lithium-ion batteries are expensive and degrade over time. Pumping water uphill to convert chemical energy to potential energy, and then releasing the water to harvest the energy, is much cheaper but requires a body of water. So, a grid operator in China found an even lower-tech solution: use surplus wind power to drive cranes that lift 35-ton concrete blocks up a 148m (486ft) tower, then drop the blocks to spin a generator whenever the grid needs electricity. The new tower is expected to have a 35-year lifespan and should enjoy 80% round-trip efficiency with no chemical degradation (or water evaporation).

🚏🏭 The US now invests more in fossil fuel energy than China does

For decades, China invested more in fossil fuel energy generation than the US did, but a combination of recent policy changes and Iran-war-related shortages has led China to cut investment, while the datacenter boom and political shifts have led the US to quintuple its annual investments in fossil fuel generation over the last few years. And so, the US has suddenly overtaken China in fossil fuel investments after decades of sitting comfortably behind. In related news, every region of the world, bar one, reduced its carbon intensity of energy (how much CO2 is emitted for a given amount of energy generated) in 2025, meaning their energy mix became greener… the lone exception being North America, which saw a substantial increase.

🚏🗽 NYC apartment construction hit a 60-year high

The United States continues to struggle with housing shortages — nationwide apartment construction recently hit a 15-year low — but New York City has been building apartments at a record pace: it added 38,682 units to its housing stock last year, the largest gain since 1965. The boom “shows no sign of slowing down,” either, with over 16,000 units proposed in the first quarter of 2026 alone. But the Big Apple’s metro area is still 400,000 homes short of meeting demand, according to Zillow, and the city’s median rent has skyrocketed since COVID.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Theory-Free Takeoff (Learning From Examples) — Argues that technological progress doesn’t necessarily require scientific breakthroughs: you can iteratively improve a technology (within a finite search space) without understanding the core mechanisms at play. Humans have been doing this kind of “theory-free search” for thousands of years, and recent improvements in frontier AI — where LLMs’ utility progresses faster than our understanding — are just the latest example.

• The Rise and Fall of Rad Power Bikes (Geekwire) — Argues that the collapse of North America’s largest electric bicycle brand was a casualty of too much growth too fast, showing that massive funding rounds can be a death sentence for hardware businesses with thin margins, not a sign of health. What’s more, like many businesses that boomed during COVID, Rad mistook a temporary spike in demand during the pandemic for a permanent, secular shift in the market.

• The Thermodynamics of Capital (Monthly Review) — Argues that, despite the common imagery of AI as a pristine digital product freed from the dirty physical world, it’s subject to the same “political economy of entropy” as any other technology: it consumes low-entropy inputs (water, power, and minerals), processes them in the material world (in the form of data centers), and outputs high-entropy waste (such as noise pollution, CO2 emissions, and e-waste). This “dissipative” model of using an entropy gradient to do work is common to many complex adaptive systems, like a living cell or a hurricane.

• California Is Chasing Wealth That Has Feet (Progress and Poverty) — Argues that the Golden State’s potential wealth tax on billionaires is likely to fail for the simple reason that it’s trivially easy for hyper-wealthy people to move out of state. Taxing land value would be better because California actually has eight times more money in its land than in the pockets of all its billionaires combined, and land (obviously) can’t get up and move. (The root problem is California’s Prop 13, which limits how much property tax the state can collect, forcing it to rely on high income taxes and now wealth taxes.)

🔍📆 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: accountability.

Well, your account was disabled. The email doesn’t say why. The appeal button leads to a form. You find and follow the phone tree and finally get to a person, just to be told, with utmost sympathy, that there’s nothing they can do. A decision was made, but you have no way to understand or remedy it. It feels like no one is accountable.

What is accountability? Often it gets reduced to a vague concept of responsibility. But we can be more specific. In his recent encyclical Magnifica Humanitas, Pope Leo XIV offers a usefully concrete definition. Accountability means identifying who must “account” for a decision: justify it, monitor it, and when needed, answer challenges and remedy the harm. This moves accountability from a vague virtue to a set of observable properties, each of which can fail independently.

And yet, these are not all equal. No amount of justification, monitoring, and challenging makes a system accountable if there isn’t room for remedy. This can happen in any system but especially prevalent when human judgment is removed from the system, be it through “the algorithm,” “the policy,” or a committee. It doesn’t take negative intent to lose accountability. The process which adds more consistency is great until it makes a technically right but obviously wrong decision.

What we can do is always ask ourselves where the accountability lies. Who can justify, monitor, challenge, and remedy? And if you can’t tell the answer, it may be time for something to change.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 241 — July 2, 2026 — Available at read.fluxcollective.org/p/241

Contributors to this issue: Neel Mehta, Boris Smus, Erika Rice Scherpelz, Dart Lindsley, MK

Additional insights from: Ade Oshineye, Ben Mathes, Jasen Robillard, Jon Lebensold, Justin Quimby, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“Not everything that is faced can be changed, but nothing can be changed until it is faced.”

— James Baldwin (1962)

🎯🚘 Point, don’t steer

Given a task to build a treehouse, would you rather have dozens of pages of detailed schematics or a simple one pager outlining what someone should feel when they are in the treehouse?

Well, like many things, it depends. Clarity is about the goal (why this treehouse matters, what good looks like for this treehouse). Meanwhile, detail focuses on what to do and how to do it (how do we build this treehouse?). The trap for many is that detail often feels like it increases clarity, but often the opposite happens. When the intent is buried under a hundred specifics, people lose the plot. They build exactly what was described without understanding whether or not it solves the problem.

The military tradition of mission command tells officers the objective and the constraints and leaves the how to the people closest to the fighting. Its opposite, detailed command, scripts the steps from the top. Mission command wins more often not because soldiers are smarter than generals, but because they see the situation on the ground.

Detail is not always the enemy. Sometimes, a single centralized logic needs to be scaled to more than one person’s worth of execution. If an engineering system is having an outage, the incident commander needs to be able to quickly make the calls and have someone execute them. There’s no time to give high level goals and iterate for alignment. A startup three weeks from insolvency is in a similar situation. These situations tend to appear when the stakes are existential and time is scarce.

But when you’re not in those situations, detailed mode becomes a liability. Every decision routes through one person, so the people around them stop exercising judgment. Which means that now the leader has to stay involved, reducing judgment even more. It’s a feedback loop that drops an organization’s throughput to that of the bottleneck. And worse, the leader might not realize that they’re the bottleneck. Instead, they blame it on the lack of autonomy of those around them.

Detail can be valuable, but we need to figure out when and how to present it. Interface design calls this progressive disclosure. First, set the clear goal. Reveal detail on demand, when someone hits a wall. A new hire will need a lot of it; tenured engineers, almost none. The detail arrives as an answer to a real question, not a preemptive script.

As we provide guidance, we can think “does this guidance provide clarity? or does it add detail?” And if it adds detail, “Was this detail requested?” We might find ourselves realizing that clarity is, in fact, the harder job.

(Credit to Matt Manela for inspiring this piece.)

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏💱 Wealth managers are moving upmarket, leaving non-millionaires to AI

A senior partner at McKinsey observed that “mass-affluent” clients, defined as those with liquid assets between $100,000 and $1 million, can now get “private-banking quality from AI,” meaning that financial advisors who focused on standardized advice for the mass market are now a lot less valuable. Instead, he predicted that wealth managers will shift to serving even richer clients, since those people often have emotional needs (such as which child will get the largest inheritance) that require a human touch. Banks such as Citigroup have actually stepped up hiring for wealth advisors, but the roles are shifting now that internal AI tools can do a lot of basic portfolio analysis. The McKinsey partner predicted that firms would need to start hiring for AI experts, “behavioral data scientists, personalization architects, and human-in-the-loop oversight professionals,” all hybrid roles that didn’t exist in wealth management just a few years ago.

🚏🎰 Influencers promoting a prediction market “won” millions on a fake betting site

A duo of exposés from Politico and the Wall Street Journal found that the prediction market startup Polymarket paid influencers to post videos of themselves winning hundreds of thousands of dollars (or more) betting on things like what a politician would say in a speech. It turned out those bets had never actually been placed, and if you’d actually placed the hundred-plus bets featured in the ads, you’d have lost $166,000. It turned out that Polymarket devs had built a fake lookalike website, “poiymarket.com” (with an ‘i’ in place of the ‘l’), for influencers to post these mock trades. What’s more, Polymarket’s CMO used his personal PayPal account to send over $2.5 million to political influencers, most of whom never disclosed that they were doing a paid partnership.

🚏🛻 A dead-simple electric truck will sell for $25,000

The average new car in the USA sells for $50,000; even used cars sell for upwards of $25,000. But an EV startup called Slate has revealed a bare-bones electric pickup truck that will sell for just $24,950 (not counting fees or taxes). The base model has hand-crank windows, basic gray paint, no infotainment system, and just 205 miles of range, but it’s modular and allows buyers to add things like roof racks, upgraded headlights, or a back section that turns it into an SUV. While the US is indeed lacking in affordable EVs, it remains to be seen if there’s market interest for a low-end pickup truck in a country where the average new pickup sells for over $66,000.

🚏🗳️ An NYC politician was arrested for forgery over AI-generated photos

A politician from Queens who ran for state assembly last year was arrested for forgery after it was revealed that he’d used AI to create fake endorsements, photos, and news articles. He made a social media post claiming the endorsement of the Queens Jewish Alliance, using their logo on a realistic looking (but AI-generated) endorsement sheet; he also swapped his face into a picture of a prominent local politician shaking a man’s hand, making it appear that that politician supported him. In the complaint documents, the Queens District Attorney added that this man had deepfaked photos of his opponent and generated AI videos “appearing to show endorsements from a police precinct and an elementary school.”

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Your AI Is Not a Tool (The Convivial Society) — Applies a McLuhanist lens to argue that AI, like most modern digital technologies, isn’t really a tool. Tools exist outside of us and channel certain skills, and they can be picked up and put down at will. But AI is an environment that envelops us and shapes us; you can’t escape it or choose not to engage. This line of thinking refutes the cliché that “technology is just a tool, and what matters is what we do with it.”

• Why Flying Within Africa is Harder Than Flying Out of It (Proxima Imperium) — Argues that the lack of competition (and resulting high prices) of intra-African flight routes is due not to corruption but to a collective action problem. Countries signed agreements to allow airlines to freely cross national borders, which would stimulate competition and drive down prices, but each individual country has an incentive to defect and lock down its airspace to prop up its (often struggling) state-owned flag carrier. Thus, good treaties like the Yamoussoukro Decision are enacted de jure, but everyone slow-walks the implementation, leaving the plans as good as useless.

• AI Isn’t Management. Try Explaining That to Matthew Prince. (Programmable Mutter) — Political scientist Henry Farrell examines a tech CEO’s “self-congratulatory op-ed” where he talked about growing his business while laying off 20% of its employees; the essay cited the writings of famed management consultant Peter Drucker to support its argument that middle managers (“measurers”) can be eliminated. Farrell argues that Prince misread Drucker; Drucker says that technology can and should be used to automate menial work and free employees up to make higher-level judgments and strategic decisions. This vision would actually lead to more mid-level management, not less, and Drucker would say it’s a good thing because it helps develop humans’ capacity and leads them to self-actualization.

• Iran’s Ultimate Banned Book (Amir Ahmadi Arian) — Examines Iranian writer Sadeq Hedayat’s 1936 masterpiece The Blind Owl, which holds the rare distinction of being banned both before and after the Iranian revolution, surviving in a form similar to Soviet samizdat. The hallucinatory, opium-fueled novel follows a pen-case painter who descends into a delirious oblivion, ultimately transforming into the very Quran-reciting old man he despises. Hedayat deeply loved his homeland yet routinely reviled it in letters as “Stinkistan.” Named for a bird that portends catastrophe rather than wisdom in Persian culture, the novel offers an absolute, uncompromising pessimism.

🔍📅 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: prognosis vs. diagnosis.

“You’ll be fine in two weeks,” is a reassuring thing for a doctor to say, but it leaves out what’s actually wrong with you. It is a prognosis — a prediction about where you end up. It says nothing about the diagnosis, which is what’s actually causing that outcome.

Physicians used to be excellent prognosticians and useless diagnosticians — and not that long ago, either. This is at first surprising, since the future is naturally less certain than the present. In 1850 a skilled doctor could often tell you whether you’d survive a fever, without the faintest idea of its source — or worse, with certainty that the fever was caused by humoral imbalance, not bacterial infection.

The two reduce different kinds of uncertainty. Since a prognosis forecasts the outcome, a good one is confident, specific, and correct (often enough). Diagnosis goes after the cause and can identify levers, directing us toward how to intervene.

Sometimes we mistake the prognosis for the diagnosis. “This project will slip three weeks” tells us something valuable, but it doesn’t say why. Maybe a handoff broke. Maybe nobody owned it. You can become a brilliant forecaster of dysfunction and remain completely unable to treat it or, even worse, you can end up treating the symptoms. Adding more status meetings or more staffing to a late project may very well be the blood letting of the organizational world. Bring me the leeches!

Next time something’s wrong, ask: Do I need a prognosis? diagnosis? both? (But, unlike with germs, when you’re working with humans, remember to engage in connection before correction.)

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 240 — June 18th, 2026 — Available at read.fluxcollective.org/p/239

Contributors to this issue: Jasen Robillard, Erika Rice Scherpelz, Ade Oshineye, Neel Mehta, Boris Smus, MK

Additional insights from: Ben Mathes, Dart Lindsley, Jon Lebensold, Justin Quimby, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“We have simply got to make people aware that none of us are free until we’re all free.”

— Opal Lee

🪂🪜 Failing from above and below

Watch two athletes training to compete as Olympic lugers. The first is patient, methodical, studious — she does as her coaches tell her and trusts the limits they set. Hers is an incremental-win approach. The second is impatient, unorthodox, “wild,” intuitive. She plays at the edges, thinks and competes outside the box, and trusts the safety systems to catch her when she overcooks a turn. Hers is an innovation/disruption approach: fail early, fail often, yet fail safely. Competing on the same ice, the two have completely different relationships to failure.

Optimal performance lives just below the surface of failure. This surface of failure isn’t a flat plane; it is a complex three-dimensional edge, one whose shape shifts with skill, mood, the day. Push right up to it and you find your best. But you’ll crash if you breach it.

The cautious athlete fails from below the surface, from the ground of known safety. She approaches that edge cautiously, often with an ingrained sense of trepidation. She can struggle to reach her best, leaving real performance on the table. This is the failure to lift off for fear of melting one’s Icarus wings.

The wild athlete fails from above. She ignores the risk or perhaps trusts the safety net. She lives by the motto: “If I’m not failing, I’m not trying hard enough.” Thus rather than fail with timidity, she fails quite spectacularly. Many of the most important lessons get learned this way, half by accident. But the risk is that when she fails, the failure might be catastrophic, ending a career or even a life. When Icarus gets too close to the sun without a failsafe, he falls to his doom.

Yet the evidence of failed flight attempts maps out a corridor of success that the rest of us can navigate through.

The wild athletes chart the boundary; the careful ones turn that knowledge into repeatable wins. Both approaches interact with the same envelope of failure. What changes is how fast, and at what cost, the athletes learn from their timid and spectacular mistakes. Both approaches benefit from failsafes which unlock recoverable failure as the engine of learning. A failsafe doesn’t prevent failure — it determines the tuition you pay when failures invariably show up.

So, let the wild thrill seekers safely play at the edge. Coach those with lower risk tolerance to push past their comfort zone to learn and grow from where the spectacular failures landed. Allow both to succeed in their way, by leaning on systems designed so that failure leaves wax and feathers on the ground, not a body in the water.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🛩️ The world’s first crewed, battery-powered plane just took off

Fixed-wing airplanes have long been seen as difficult to electrify, since weight is everything in the air and lithium-ion batteries don’t have nearly the same energy density as jet fuel (to say nothing of their flammability problem.) But one startup has created solid-state batteries that hit 60% more power per kilogram than their previous lithium-ion pack; they plugged those into their prototype plane for a test flight this month in Florida, making it “the first crewed, fixed-wing aircraft ever to fly on solid-state batteries.” The plane, which started life as a motorized glider, can also recharge mid-flight with wing-mounted solar panels and a turbine that generates power from wind.

🚏🤠 Austin’s housing boom drove down rents despite the city’s rapid growth

Austin, Texas has been booming since the 2000s, but it came at a cost: rents rose 93% from 2010 to 2019, the most of any major American city, and home prices rose 82%, more than any other Texas metro. But a slate of zoning reforms in 2015 (such as removing parking minimums and enabling ADU construction), plus a huge bond measure to build affordable housing, led Austin to add 120,000 housing units from 2015 to 2024. That’s an increase of 30%, more than triple the US’s overall housing growth rate of 9%. The increased supply has been great news for renters: Austin’s median rent fell from $1,546 per month in 2021 to $1,296 this year, putting it below the US’s nationwide median of $1,353.

🚏🧸 The UK will ban social media for children under 16

Following an Australian law that bans children under 16 from using social media, the UK has announced a similar rule that will bar under-16s from social platforms like Instagram, YouTube, TikTok, and Snapchat, though messaging apps like WhatsApp will be exempt. The move comes amid concerns over excessive screen time’s impact on kids’ well-being and the mental health effects of social media; one British poll found that 90% of parents supported a ban like this. However, enforcement has proven difficult, as kids often find ways around age verification, video selfies with fake mustaches or AI-generated videos; 70% of Aussie parents said their kids still use social media despite the ban down under. There are also significant privacy and security risks when tech companies need to gather people’s IDs or facial scans for mandatory age verification.

🚏🍏 Apple will raise phone prices amid the memory crisis

The price of memory chips has been skyrocketing due to the AI boom, and it’s been slamming the smartphone market hard: global smartphone sales are projected to decline 15% this year and a research firm predicts that phones’ average selling price will rise 20% this year. While Apple may have been resistant to passing along increased costs to consumers, outgoing CEO Tim Cook said that price hikes are now “inevitable” as the memory situation has become “unstable.” He didn’t specify what the exact price rises would be, but analysts predict that the next-gen iPhones will cost $150 more than the current iPhone 17s.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• Technology Weakens Our Minds. We Can Fix This. (The New York Times) — Cal Newport, author of “Deep Work,” argues that our current attention crisis requires a cognitive fitness revolution, drawing a parallel to how the 1970s jogging/aerobics boom transformed a sedentary American physical culture into a more active one. Algorithmic feeds, even non-ad-driven decentralized ones, run on the same rapid-fire, fragmented, high-dopamine feedback loops that are the cognitive equivalent of junk foods. What would a gym for the mind look like?

• Asymmetry of Verification and Verifier’s Rule (Jason Wei) — Argues that machine learning’s ability to optimize anything that can be measured means that AIs will be most successful solving problems whose solutions are easy to verify. In other words, there’s a single success metric that everyone agrees on, and any given solution can be verified quickly, cheaply, and in parallel. Beyond computation, look for AI to excel in spaces with easily-verified solutions, like “finding the best catalyst to speed up a chemical reaction” or picking “the best aerodynamic car design for the fastest quarter-mile time.”

• Why Do Commercial Spaces Sit Vacant? (Strong Towns) — Examines how the unusual financing of commercial real estate loans leads to unintuitive behavior: namely, how commercial landlords would rather let storefronts sit empty than rent them out at a discount. Specifically, the value of a commercial building is determined by its going rent, so if you cut the rent, you may become underwater on your loan. Better to “extend and pretend” to maintain your building’s paper value.

🌀🖋 More from FLUXers

Highlighting independent publications from FLUX contributors.

In “Diamond-Complete: The Shape of Work That Compounds with AI,” FLUX’s own Anthea Roberts takes a fresh look at the “T-shaped” metaphor for skills, where someone has both broad generalist knowledge (a horizontal line) and deep knowledge of one particular field (the downward line).

Anthea writes that in the age of AI, horizontal generalists are at risk of being replaced because LLMs (which have a decent, if flawed, understanding of every topic under the sun) can do their jobs; vertical specialists have a domain-specific edge on AI, but it erodes every year as LLMs get smarter. Being T-shaped helps, but the best move is to gain AI proficiency, which in this metaphor adds an upward line to the T, forming a perfect plus-sign and making oneself “diamond-complete.”

This blend of AI fluency, general knowledge, and domain expertise allows you to get the most out of AI with a mindset of, “I know how to direct AI to get better results than it produces by default, and I know where the model fails in my corner, and I catch it before it costs anything.” Developing diamond-shaped talent in an organization is highly difficult, though, and we can’t “let AI quietly eat the apprenticeship”: keep developing juniors who can develop the “downward line” that’s perhaps the hardest part of the diamond to learn.

🔍📅 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: transaction costs.

You break a glass and someone offers to help clean it up. But you don’t really want their help. Still, it feels a bit callous to reject help sincerely given.

But help does not come without cost. All exchanges, even social exchanges, come with a transaction cost. Exchange creates overhead, and sometimes that overhead can be worth more than the exchange itself.

For example, the offer of “help” can create work, as anyone who has gotten help from a toddler can attest (we let the toddler help for other reasons, bonding, education… not the value of their output). Getting help creates a cost of delegation: you have to explain the task and review the output. Especially if you’re still doing part of the work, the mental energy of doing this can be more than having just done it yourself.

There’s a sneaky type of transaction cost that comes when an exchange, once offered, can’t be declined for free. These sorts of costs are especially prevalent in social situations, whether it’s the potential ingratitude of refusing sincerely offered help or the relationship-changing fallout of refusing a marriage proposal (“we can just keep dating, right?” rarely goes over well).

It’s not that we should think of everything through a cost/benefit lens. But when a reaction — yours or someone else’s — feels a bit too strong for the situation… it’s worth taking a look at what hidden transaction costs may be in play.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 239 — June 16th, 2026 — Available at read.fluxcollective.org/p/239

Contributors to this issue: Neel Mehta, Justin Quimby, Jon Lebensold, Erika Rice Scherpelz, MK

Additional insights from: Ade Oshineye, Ben Mathes, Boris Smus, Dart Lindsley, Jasen Robillard, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“In a fully developed bureaucracy there is nobody left with whom one can argue, to whom one can present grievances, on whom the pressures of power can be exerted. Bureaucracy is the form of government in which everybody is deprived of political freedom, of the power to act; for the rule by Nobody is not no-rule, and where all are equally powerless, we have a tyranny without a tyrant.”

— Hannah Arendt, On Violence

👑🖥️ The sovereign’s prompt

For most of history, the ability to command another person was a mark of status. The powerful spoke through intermediaries: scribes, secretaries, amanuenses, chiefs of staff. The privilege of the ruler was that they did not have to descend into the details.

AI agents semi-democratize that privilege. Now almost anyone can have something like a tiny administrative team. When we prompt, we occupy the position of command. (And in The prompt test we noticed how we’re often more patient with these virtual assistants than we are with real people we delegate to.)

But something has changed: who is responsible? When a person gives us the wrong result, we tend to treat the failure as theirs (even if we underspecified the task). In fact, that asymmetry is one of the oldest perks of command. An oft-ignored element of delegating to a person is that if they messed up, it was their neck on the line. And if we, as sovereign, messed up... well, it was often still their neck on the line.

If our agent sends the wrong message, books the wrong trip, or builds the wrong feature, whose judgment was actually operating? Ours? The model’s? The tool that wrapped the model? The prompt starts to look less like a request and more like a chain of command, one where the accountability is vague.

Accountability is not the only thing we lose in this model. If everyone becomes sovereign of their own little kingdom, what happens to the many unstated roles that intermediaries used to play? Intermediaries can translate for the target audience. They can distribute institutional memory. They can add the kind of friction that’s useful, such as making sure change doesn’t happen too quickly for the organization to absorb. Intermediaries can also apply additional judgment, such as when the sovereign’s idea isn’t one that should quite be taken literally. Some of these roles may move into the agents, some may move to the supervisor. But the key thing to note is that the people who “just” executed commands did a lot more than bare execution.

The sovereign’s prompt is not just a better instruction. It is a political object. It reveals what was hidden before: who gets to be vague and who has to be precise, who gets to iterate and who gets blamed.

So as we rebuild our world around AI-based automation, think about what becomes explicit, and the costs and benefits of doing so.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🌯 A meme coding agent runs on “stolen compute” from customer support bots

Chipotle’s customer support chatbot went viral after people realized it could answer coding questions for them. So, some hackers reverse-engineered how to connect to Chipotle’s backend to get free AI calls and plugged it into OpenCode, a popular open-source coding agent. The resulting app, dubbed Chipotlai, thus allows developers to run coding agents for free using “stolen compute” from big companies’ support bots. They’ve since expanded from Chipotle to Lowe’s, the Home Depot, Expedia, IKEA, and more.

🚏🇲🇽 The Mexican government is developing a cheap electric car

The Mexican President recently unveiled a prototype of the Olinia Uno, a planned six-seat electric car that would sell for just $8,500; it’s made in Mexico using primarily domestic parts. The Mexican government invested in the car’s development as part of an initiative to boost the country’s manufacturing sector, and it’ll also install 2,000 charging stations across Mexico City and nearby states. The move comes at a good time, as Mexico’s EV market is booming: EVs account for 9.5% of new vehicle sales in the country, and Mexico is now the leading electric car supplier to the US.

🚏🌨️ Cloudflare says bot traffic now exceeds human traffic on the internet

Cloudflare’s CEO posted on social media that “agentic traffic [is] growing so fast that bots have now passed human traffic online for the first time in the Internet’s history.” Bots now make up about 57% of requests for HTML webpages, though many of these are likely quick, high-volume page scrapes rather than purposeful browsing (meaning that humans probably still dominate in terms of time spent on the web). Certain small regions like Gibraltar and Singapore seem to be hotspots for agents (whether via VPNs or data centers), with over 90% of web activity in those places coming from bots, and most traffic coming from desktop and requesting structured JSON data, a telltale sign of calling APIs.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• My Prodigal Brainchild (Neal Stephenson) — The author who coined the term “metaverse” argues that the metaverse has been created in the form of Minecraft, Roblox, Fortnite, et al., and in fact it was inevitable as soon as we got interconnected computers that could show graphics and track persistent game state. However, the 2021-era vision of a metaverse mediated by VR headsets was just one (highly flawed) implementation, and its collapse doesn’t mean the metaverse overall is dead.

• Study Solves the Mystery of Why Humans Are Mostly Right Handed (Anton Petrov / YouTube) — Finds that humans, along with most hominins, are primarily right-handed because the left brain (which controls the right side of the body) was ideal for complex tool making, which requires fine motor control and mastery of repetitive, sequential steps. The right hand was better at tool use and therefore became preferred in ancient humans. Interestingly, the “fighting hypothesis” posits that a small fraction of lefties stayed around because they’d get a surprise advantage in combat against right-handed opponents.

• The A.I. Fear Keeping Silicon Valley Up at Night (New York Times) — Argues that centralized AI risks consolidating power into a corporate-state feedback loop, creating a “permanent underclass” that exposes the precarity of the tech worker class itself. The antidote may be to reclaim cognitive sovereignty by running local AI models to bypass centralized oligopolies and unlock decentralized production.

• The 5 Biggest Obstacles to AI Data Centers in Space (Big Think) — Argues that the much-ballyhooed concept of data centers in space may not live up to the hype. Costs go down and engineering breakthroughs can be achieved, but the laws of physics get the last laugh: cooling is extremely tough in a vacuum where all you have is radiation, and cosmic ray bit-flips are inevitable but difficult to detect or repair.

💊🍀 A dose of hopepunk

An optimistic sign that our world’s systems are changing for the better.

After centuries of rampant growth, carbon emissions are finally starting to plateau: “after increasing by more than 20% in the 2000s, CO₂ emissions today are a mere 3% higher than they were in 2013.” Major drivers include the growth of renewable energy and, less obviously, reductions in deforestation. Atmospheric CO₂ now sits at 431ppm, well below the earlier “status quo” projections of 439ppm, but it’s still increasing linearly because we still emit more carbon than the planet can absorb. (There’s also an interesting, and unfortunate, second-level feedback loop to consider: carbon sinks may get less effective as the planet warms.)

🧩🕹️ Systemic tools and toys

Games, gadgets, demos, and other nifty interactives to stretch your systems-thinking capacity.

Resource management, long-term planning, and calculated risk-taking are key in a quintet of sports web games where you have to assemble a team of historical players who would go undefeated in a simulation season. There’s the original 82-0 for basketball, 20-0 for American football, 8-0 for soccer (“normal” football), 162-0 for baseball, and 98-0 for ice hockey.

In the first round you can pick any player from a randomly chosen team’s roster, but at the end you only have one position left to fill, so strategic routing is essential: you don’t want to leave a tough position for the last round and be left picking between a pile of duds! For instance, in the baseball game (where FLUX’s high score is 158-4), most teams have a hard-hitting first baseman, so you could leave that slot for the end, whereas decent catchers are very rare, so you want to prioritize grabbing a solid one as soon as possible.

Leaving shortstops to the end was a mistake! See if you can beat us.

🔮📬 Postcard from the future

// What might letters written in the style of US Civil War correspondence during the AI push of 2026 look like?

“Dearest spouse, it is hard to see the glory of our company’s market dominance through the mighty chugging layoff machine, trampling the laid off beneath its cruelest algorithms. Communications cut off in the middle of the night. No last happy hour or lunchtime walk to speak a word of comfort to those impacted. Heaven grant there may be an end soon.”

“Mother and Father, thank you for the recent correspondence and images of you at home. That fixed look of care and sorrow upon your faces fills my soul with the keenest anguish. I would most gladly give all my hopes of happiness to save thee from one moment of pain or grief. But that can not be for the Board has deemed it otherwise, as I must fight with tooth and nail for compute to train my models.”

“I joined my onboarding class some nine months ago with fellow idealistic and driven people. We now number only three. H.J.F. was well when I last heard from him about a month ago from within the maws of the data center. I have not heard from Todd since I left the last all-hands. He is probably in legal review with the dreaded EU compliance teams.”

“Professor Q, I find myself as a cork tossed about on a storm-swept ocean. The lofty goals of greater glory for humanity we wrote about seem to be traded by my leaders for sacks of gold. ‘We must top the frontier model charts’ they cry, whilst urging the bending and skirting of moral and social guides in a quest for context to train upon. I fear the writings of my youth, nigh five years ago, reveal me an idealistic fool, ignorant of the ways of the world.”

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

June 4th, 2026Episode 238 — June 4th, 2026 — Available at read.fluxcollective.org/p/238

Contributors to this issue: Ade Oshineye, Ben Mathes, Erika Rice Scherpelz, Dart Lindsley, Justin Quimby, Neel Mehta, Boris Smus, MK

Additional insights from: Jasen Robillard, Jon Lebensold, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“The prophecy of doom is to be given greater heed than the prophecy of bliss.”

— Hans Jonas

🎲👁️ When the odds aren’t yours

You’re in front of a door that sends nine out of ten people to Disneyland and the remaining one to Mordor. Do you go through?

It looks like a pretty good bet, most likely a day of leisure topped off by fireworks. Most people won’t have to deal with orcs and giant spiders (and the slight inconvenience of probable death). But if you are the one who ends up in Mordor, the average is not terribly consoling. The upside is pleasant. The downside is ruin.

In general, the group average can look fine while the individual path through the system is not (see: non-ergodicity). Push this to the extreme, and we can only afford to get it wrong once. And it doesn’t matter if the odds of unbearable loss are rare when they are unpredictable.

Moreover, when harm is irregularly distributed, internally rational responses can look to observers paranoid or oversensitive. The constant defense of decisions and lived reality ends up becoming a full time role in the management of other people’s skepticism. Over time, we learn to quietly route around danger rather than increasing the legibility of a harmful system to those who are spared its costs.

Asymmetric outcomes are a consequence of how complex systems distribute risk unevenly. So, before judging a decision, we can ask, “Who owns the downside?” When we have some control over the system, we can find ways to distribute the upside and downside more symmetrically. Rather than automatically assuming they’re character flaws, consider that caution and “overreaction” may be signals that we’re missing something.

When someone refuses the door, don’t ask, “Why are they so afraid of Disneyland?” Ask, “What do they know about Mordor?”

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏💎 Lab-grown diamonds are the latest beneficiary of the AI boom

Synthetic diamonds are the latest unexpected product to see a surge in demand due to the AI, semiconductor, and datacenter boom; the share prices of several Chinese lab-grown diamond firms have skyrocketed in the last few months. Diamonds are exceptionally good at conducting heat but poor at conducting electricity, which makes them useful ‘heat spreaders’ for silicon chips, enabling denser and thus more powerful chips.

🚏🚀 Nasdaq relaxed its index listing rules right before SpaceX’s IPO

Traditionally, tech stocks would have to wait up to a year between going public and getting listed on the Nasdaq 100, the list of the hundred largest non-financial companies listed on the stock exchange. This rule was designed to give time for price discovery and let the stock price settle; sidestep the post-IPO slump that many stocks experience; stop hedge funds with early access to IPOs from frontrunning investors; and let insiders with 90- or 180-day lockups sell before passive investors in index funds had to buy the stock. But the stock market recently changed the rules to allow companies to enter just 15 days after IPO, a move that was apparently designed to lure SpaceX to IPO on Nasdaq instead of the rival NYSE. And now, as SpaceX looks to launch the biggest IPO in history, many investment advisors view this as a “shameless structural manipulation,” as index investors will be forced to buy potentially overvalued SpaceX stock, turning their 401(k)s into “exit liquidity” for insiders.

🚏🚜 Demand is booming for a “repairable, no-tech tractor”

John Deere’s tractors, harvesters, and other agricultural machines are infamously configured so end users can’t repair them; built-in software will brick the machine unless repairs are done at an authorized dealership. This adds both delays and expense for urgent fixes. The resulting demand for a user-repairable tractor birthed a Canadian tractor company that’s developed a new tractor that’s intentionally “built without computer controls.” Demand for the machine has reportedly been through the roof, with over a thousand farmers from 30 countries writing in for information after the tractor was presented at a farm show.

🚏🚗 An American bill would ban Chinese cars from driving into the US

Two lawmakers from Michigan recently introduced a bill that would ban cars designed or manufactured in China from driving into the US from Mexico or Canada; they called it a “national security issue” because these internet-connected cars could allegedly “collect full-motion video” or “map sensitive infrastructure sites.” (This likely plays well politically in Michigan, where many auto workers feel threatened by Chinese cars.) Meanwhile, cheap Chinese cars are everywhere in Mexico, with BYD controlling 70% of Mexico’s electric car market, while Canada is greatly relaxing its tariffs on EV imports from China and will allow 49,000 Chinese EVs to enter the country this year.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• The Platonic Case Against AI Slop (Palladium) — Uses Plato’s imitation hierarchy and model collapse research to argue that slop doesn’t just lower the quality bar, it trains your soul to prefer lower bars. The antidote isn’t to avoid AI altogether, but to avoid engagement-maximizing feeds and curating actively.

• Consumer Sentiment Isn’t Politics, but Politics Is Reshaping It (Claudia Sahm) — A renowned economist observes that Michigan’s famous consumer sentiment survey, which measures people’s self-reported satisfaction with the economy, has become largely correlated with people’s political leanings: many people say they’re happy when their party controls the Presidency but instantly switch as soon as the other side takes over. (Goodhart’s Law may be at work, since the media uses consumer sentiment stats as a key indicator for the President’s success or failure.) There’s still some signal in the noise, though, such as a secular downward trend in sentiment since 2020, which matches the “vibecession” phenomenon.

• AI Is Killing the Cheap Smartphone (David Oks) — A deep-dive into the economics of the semiconductor memory industry. AI servers need a special kind of memory chip (HBM), rather than the DDR and LPDDR that phones and laptops use, so soaring AI demand has driven memory manufacturers to reallocate lines toward HBM and away from consumer memory, which has driven memory shortages and price hikes in consumer devices, thus rendering many cheap phones “permanently uneconomical.” What’s more, a history of getting punished for overbuilding supply in this cyclical industry has driven manufacturers to purposely restrain themselves from building new factories, which means we can’t simply build our way out of this shortage.

• How the US Eradicated Screwworm (Sarah Taber / Bluesky) — Details how the US worked with Central American governments to eradicate flesh-eating screwworms (a scourge of livestock farmers) using the clever method of releasing hordes of sterile male flies that would lead females to lay unviable eggs. However, the US cancelled a screwworm monitoring program last year, and the dangerous bugs have returned to Texas and are threatening livestock there.

🔍📆 Lens of the week

Introducing new ways to see the world and new tools to add to your mental arsenal.

This week’s lens: lens stacking.

Oh no! Your launch is about to slip. Engineering sees a dependency problem. Design sees low quality user experiences. Sales sees a broken business model. Finance sees burn. Each account is true... enough. None is the whole situation.

Each perspective has a lens, and lensical thinking is the practice of noticing that we all see through lenses. Lenses are not neutral, and sometimes they can get stuck to our eyeballs.

Lens stacking is what happens when we deliberately put on more than one lens without collapsing them into a single answer. Instead of asking “who is wrong?” we ask “how might we all be right?”

Each lens adds a layer of information. Like the lenses in a telescope, the stack creates a view that no lens can produce on its own. And this can happen even when the lenses don’t agree. If one says “move faster” and another says “slow down,” the tension may reveal the shape of the problem.

Too much stacking can get us stuck in analysis paralysis, or it can lead us to think we agree when really we’re just being too general. Putting on and taking off both matter. Putting another lens on the stack acknowledges complexity. Taking one off can give us more freedom to act.

Complex systems are too rich for single-lens certainty. Stack the lenses up and we can see farther.

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 237 — May 28th, 2026 — Available at read.fluxcollective.org/p/237

Contributors to this issue: Justin Quimby, Jon Lebensold, Erika Rice Scherpelz, Neel Mehta, Boris Smus

Additional insights from: Ade Oshineye, Ben Mathes, Dart Lindsley, Jasen Robillard, Lisie Lillianfeld, MK, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“A bad system will beat a good person every time.”

— W. Edwards Deming

💩🔁 Shit work begets shit work

A critical piece of bespoke software has stopped working; a consultant is hired to patch it up ASAP. There is no architecture review, no hard questions asked. Leadership needs it running again. The consultant delivers beautifully: the project was on time and they even fixed some bugs along the way. But three months later, a minor dependency update breaks it all over again. It’s not a great outcome for the client, but at least the consultant has some nice repeat business.

In other words: shit work begets shit work.

When someone expresses satisfaction with our work, they engage in a collaborative relationship calibration: they indicate what they think “good” means (or, at least, good enough) and this feedback articulates the level of care needed to get rewarded. Consequently, new requests arrive already shaped by that lesson.

This dynamic rhymes with Gresham’s Law of “bad money drives out good.” When two coins circulate at the same face value, people spend the weaker and hoard the stronger. While the bad stays visible, the good disappears from use.

Work quality also rhymes with this: if our judges (boss, client, audience) are satisfied by our work, we stay employed. If they are satisfied by mediocre work, we stay mediocre. Those demanding a higher level of craft, they will have to look elsewhere. Nobody has to complain for the system to sort itself out. Meanwhile, our audience narrows around the level of quality we have collaboratively normalized.

Skills follow the same loop. Work done well emits an aura that is more than the functional output. The thinking we skip becomes less fluent. Care, not practiced, gets harder to recapture. And the less often we’re asked for excellence, the fewer chances we get to exercise it.

Work done with unusual care attracts colleagues and clients who value care. They ask better questions. They notice better tradeoffs. Their expectations become a constraint that helps us stay good.

Not every job can be a masterpiece. Budgets are real; deadlines are real. Sometimes “good enough” is good enough. But good enough is a choice relative to a context: good enough for someone, under some set of expectations, with some future implied.

So don’t just ask, “Did this work pass muster?” Ask, “What kind of future work does it select for?”

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏🥩 Texan steakhouses are going out of business as beef prices skyrocket

Steak prices have risen 17% in the last year and ground beef prices have risen 19%, hitting a record high. In Texas, the price of wholesale brisket meat has risen 28% year-over-year. (Major drivers include tariffs; inflation; and the smallest national cattle herd in 75 years due to droughts, labor shortages, and loss of ranching land.) Barbecue joints and steakhouses have had to raise prices, limit brisket sales to just one day a week, or even shutter entirely. Meanwhile, lower-end chains like Outback Steakhouse and Texas Roadhouse, which use cheaper beef cuts and alternate meats, have shown strong revenue growth. It also helps that price-conscious consumers are looking for cheaper menu items.

🚏🌸 Wind and solar now generate more power than natural gas globally

Wind and solar power together produced 22% of the world’s electricity this April, compared to natural gas’s 20%. (Both remain far behind coal, though.) It’s the first time that the two renewables have eclipsed gas, and given their rapid growth over the past decade, they seem poised to maintain and expand their lead. It’s not surprising that this milestone first occurred in April, as the month’s mild spring weather in the Northern Hemisphere is typically good for both wind and solar, while the reduced need for heating and cooling reduces demand for gas.

🚏🇳🇱 The Netherlands blocked a US company’s takeover of a key government tech vendor

The Dutch company Solvinity runs the country’s digital ID app, which residents use to “authenticate themselves online… to book a doctor’s appointment, buy a house, or interact with public authorities.” Kyndryl, an American firm spun off from IBM, announced it was going to acquire Solvinity last year, but the Dutch government recently intervened to block the purchase, since foreign ownership of such a major piece of digital infrastructure could pose a “risk to the public interest.” It’s the latest in Europe’s tech sovereignty initiative, where the EU and individual countries across the region are attempting to reduce their reliance on American cloud, AI, and software vendors.

🚏🚕 Uber and Lyft drivers in MA formed the country’s first rideshare drivers’ union

Rideshare drivers in the US are classified as independent contractors, not traditional employees, so they don’t normally have the right to form unions. But a 2024 Massachusetts law created a special framework to allow them to organize if they could get signatures from 25% of the state’s active drivers. The newly formed App Drivers Union recently met that threshold and has now received certification from the Massachusetts Department of Labor Relations to represent the New England state’s nearly 70,000 Uber and Lyft drivers, making it the country’s first official union representing rideshare drivers.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• I Think Anthropic and OpenAI Have Found Product-Market Fit (Simon Willison) — Observes that enterprise coding agents have been able to successfully ramp up prices without losing customers (a great sign of PMF), whereas consumer AI products still need subsidies and struggle with monetization. This shift to enterprise is especially apparent if you look at the big labs’ open job listings: almost a third of them are for ‘enterprise-y’ roles like account executives, forward-deployed engineers, and go-to-market people.

• Inflation Inequality Meets Wage Compression (Arin Dube) — The economics professor observes that the post-pandemic recovery was unusual in that wages at the bottom grew more than those at the top, leading to a decrease in income inequality (“wage compression”). However, inflation was especially high in categories that poorer people spend more money on, such as food, rent, and energy. This counteracted about 15% of the relative gains at the bottom, suggesting that supply-shock inflation (the kind we experienced post-pandemic) “poses a serious risk to broad-based prosperity.”

• AI Fatigue Is Real and Nobody Talks About It (Siddhant Khare) — Argues that the Jevons paradox of AI carries a heavy psychological toll: as task times shrink, capacity expectations expand, forcing us to constantly take on more work. The critical countermeasure is protecting dedicated no-AI time to firmly separate deep human thinking from AI-assisted execution.

• Size and Dynastic Decline: The Principal-Agent Problem in Late Imperial China (Explorations in Economic History) — Develops a model to argue that Qing China’s decline in the 19th century was driven largely by a principal-agent problem in tax collecting. The emperor was generally a benevolent dictator whose long time horizon led him to set low tax rates to ensure regime stability in the long run; however, he had to defer to regional tax collectors to observe local economic conditions and collect the appropriate amount of funds, especially in the hinterlands. But these collectors had the short-term incentive to extort taxpayers and keep the excess funds for themselves. This forced the ruler to keep lowering tax rates, ultimately weakening the state’s fiscal capacity.

🍯🧠 From the hive mind

Showcasing useful tools, resources, and essays from the FLUX community.

We all know the broad story about work-from-home: a huge spike at the start of COVID, followed by a gradual decline. But once you dig into the data, there are some wild local patterns. Sunnyvale, California, had the largest percentage increase in WFH in the entire country (1,085%) from 2019 to 2021, and then the largest percentage decrease (-67%) between 2021 and 2024. It’s the epitome of the tech industry boomerang: massive early adoption, then a sharp pullback. Meanwhile, the decline has been much more muted a few counties over in Santa Rosa and Sacramento, and WFH has actually grown in the bedroom city of Stockton.

Check out the stats in your area in a new tool by data explorer Ari Lamstein.

🔮📬 Postcard from the future

A ‘what if’ piece of speculative fiction about a possible future that could result from the systemic forces changing our world.

// Torches flicker in the castle room, illuminating three figures seated around a table.

The Dwarf slams his tankard on the table. “This is ridiculous. They can’t do that to us. We’ve been together for two years, and now this!” His other hand gestures at the tome before them.

The Elf raises a single elegant eyebrow to regard the Dwarf. “Look, we knew things might change as we progressed. We agreed to adjustments. I admit, I was not expecting them to be this… dramatic.” A lithe hand rests on their own copy of the tome, a finger tapping its cover.

The Dwarf pauses only briefly to take a slug of ale, then continues, “Years of building our skills, building trust, creating a shared vision for the future. Now this! Introducing a new core stat for our characters. Which gets set by a roll of the dice that we have no influence over!”

“Oh it is worse than that, friend,” the Draconian says. “This new stat has a direct influence over all the key abilities of characters. My build is now effectively useless. My entire progression arc is going to hit a wall. I might as well reroll.” Smoke exits from its nostrils, tracing a downward path over the table edge onto the floor.

All three snap their heads as the Wizard returns to the room and takes their seat behind the gamemaster screen. They do not need to cast Detect Thoughts to read the room.

“Look, we all agreed to do an accelerated timeline with a minimum of metagaming. That means occasionally massive systemic changes occur. What used to be successful to get experience points and accumulate wealth won’t always work. You just have to decide if you want to keep your current character and adapt; create a new character, but at a much lower level; or drop out entirely. The choice is yours.” The Wizard rises to leave, their rulebooks levitating to follow behind them.

The players look at their tomes with resignation. “A Summoner’s Guide to the Chatbox: Mundanes & Mortgages Expansion Pack.”

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.

Episode 236 — May 21st, 2026 — Available at read.fluxcollective.org/p/236

Contributors to this issue: Justin Quimby, Erika Rice Scherpelz, Ade Oshineye, Jon Lebensold, MK, Jasen Robillard, Neel Mehta, Boris Smus

Additional insights from: Ben Mathes, Dart Lindsley, Lisie Lillianfeld, Robinson Eaton, Spencer Pitman, Stefano Mazzocchi, Wesley Beary, and the rest of the FLUX Collective

We’re a ragtag band of systems thinkers who have been dedicating our early mornings to finding new lenses to help you make sense of the complex world we live in. This newsletter is a collection of patterns we’ve noticed in recent weeks.

“A system as a whole is not an object but a way of looking at an object. It focuses on some holistic phenomenon which can only be understood as a product of interaction among parts.”

— Christopher Alexander

👔♟️The games we play

In a world where most can afford decent shoes, function becomes table stakes. Whether $20 or $200, the shoes still protect your feet. And with the baseline met, the details start carrying social meaning: the brand, the colorway, whether they’re perfectly clean or intentionally worn down. Some might say none of this matters. Materially, they’re right. If the shoe fits, wear it. But people do care. Shoes can signal taste, tribe, status, wealth, age, even ambition.

Status games don’t disappear once basic needs are met. They migrate upward. As societies become more materially abundant, people spend less energy distinguishing themselves through access to things that aid survival (more calories, higher quality housing). Instead, that energy goes to distinguishing their lives through identity, aesthetics, and cultural fluency.

So status games always exist. But in a post-scarcity world, you have more choice about which status games you play… hopefully. In some systems, opting out of the status game locks you out of jobs, relationships, or economic security — the traditional stereotype of the lonely high school nerd.

In others, it mostly locks you out of prestige within a particular social scene, but you can often find another to land in — are you a comic book fanatic? A boardgame geek? A chess nerd? There’s a profound difference between a society where status hierarchies determine survival and one where they are a tool for identity and community. (Although both can be harmful for those at the bottom of the hierarchy.)

This is where the distinction between intrinsic and extrinsic status games becomes useful. Extrinsic games revolve around recognition and social legibility. They reward signaling, performance, and visibility within a network. Intrinsic games emerge from direct engagement: becoming more capable, building something durable, developing passion or judgment. This isn’t quite the same distinction as opt-in vs. opt-out games, there’s a correlation.

Even the most abundant society cannot eliminate status competition. But we get to choose which games we want to play. You can check out anytime you like, but you can never leave.

🛣️🚩 Signposts

Clues that point to where our changing world might lead us.

🚏⛓️‍💥 Iran wants to charge fees for internet cables under the Strait of Hormuz

Fresh off a recent plan to “establish a permanent toll system” for boats transiting the Strait of Hormuz, Iran now wants to “impose fees on internet cables” running along the bottom of the Strait, with the implied threat of cutting the cables (or at least hindering repair boats) if American tech giants don’t pay up. Only 1% of global internet traffic runs through the Strait, but experts warn that cutting cables here could slow financial transactions between Europe and Asia or even drive internet blackouts in East Africa.

A map of subsea internet cables in the Persian Gulf. Source: Submarine Cable Map

🚏🎰 Minnesota became the first state to ban prediction markets

A new Minnesota law makes it a crime to host or advertise a prediction market, defined as “a system that allows consumers to place a wager on the future outcome of a specified event that is not determined or affected by the performance of the parties to the contract,” within the state. (There’s a small carveout for weather futures, which have non-gambling utility for farmers.) Startups like Kalshi and Polymarket will have to leave the state by August or risk facing felony charges. The CFTC, the federal agency that regulates prediction markets, has sued Minnesota — along with several other states that have attempted to regulate prediction markets — to try to block the bill, saying that only the CFTC should have jurisdiction over these markets.

🚏🔬 arXiv will ban researchers for a year if they submit LLM-generated papers

The online preprint repository arXiv allows researchers to upload papers before they’re peer-reviewed, making it a popular hub for free scientific literature. It has reportedly been getting spammed by low-quality, AI-generated papers: last year it announced it would stop accepting all computer science “review articles and position papers” (though novel research was still allowed). Later, it started requiring first-time posters to get an endorsement from an established scientist. The organization recently enacted an even stricter rule: anyone whose submission has “incontrovertible evidence” of LLM generation will be banned from uploading for a year. Such evidence includes “hallucinated references” or chat output from LLMs like “would you like me to make any changes?” or “fill in the above table with real numbers from your experiments.”

🚏⚜️ New Orleans could be “surrounded by the Gulf of Mexico” by 2100

A new study found that Southern Louisiana could be facing 3 to 7 meters of sea level rise by 2100, which would leave New Orleans as a mere island in the Gulf of Mexico in the best case, or completely submerged in the worst case. This dire scenario makes coastal Louisiana perhaps “the most physically vulnerable coastal zone in the world”; it has “evidently already crossed the path of no return.” The study’s authors urged the government to prepare for a managed retreat from NOLA and relocate the city’s inhabitants to higher ground.

📖⏳ Worth your time

Some especially insightful pieces we’ve read, watched, and listened to recently.

• What Will Be Scarce? (Alex Imas) — An economics professor argues that economic revolutions like AI make certain goods (and, in this case, services) cheap and abundant, but scarcity still exists; it just shifts, and that reshapes the balance of employment between sectors. In AI’s case, the services sector may become far more efficient, cutting employment there, but things with a human touch will remain scarce and highly demanded, since these feed people’s desires for exclusivity and status (high up on Maslow’s pyramid). Thus, we may see rapid growth of a “relational sector” of the economy, focused on jobs where “the human element is the product,” such as in hospitality, coaching, childcare, live entertainment, and tourism.

• Pax Electricana (Energy and Stuff) — Argues that the global economy wastes massive tonnage shipping raw fossil fuels to process raw ores; instead, we should refine materials locally using renewable energy and ship the finished goods (“embedded clean energy”). Because electricity cannot easily cross oceans, refining raw materials at the source is the only practical way to export a country’s clean energy surplus; it has the pleasant side effect of cutting global shipping weight by two-thirds and shifting geopolitical power toward manufactured abundance.

• How to Walk Through Walls (Henrik Karlsson) — Building on last week’s “hacker mindset” piece, the author explores how one can develop this mindset of understanding how the atomic machinery of a system actually works and manipulating it at the low level, without getting stuck at the high-level abstractions. As an example, consider the self-trained filmmaker Robert Rodriguez, who broke all the conventional rules of filmmaking when he spent a mere $7,000 to cobble together his film El Mariachi, which ended up grossing millions of dollars internationally.

• The “Average” Words for Things (Nardi / YouTube) — While linguists normally think that the relationship between words’ sounds and meanings is arbitrary, certain words have unusual sound patterns across unrelated languages. Many times it’s due to sensory relationships; for instance, many languages use the labial sounds /m/ or /b/ for “mouth” or the nasal sound /n/ for “nose.” This creator made an interactive site called Phonaesthesia that looks for these cross-linguistic patterns and computes an “average” word for each concept across hundreds of languages.

🔮📬 Postcard from the Future

A ‘what if’ piece of speculative fiction about a possible future that could result from the systemic forces changing other worlds.

// What does an oncology convention sound like in 2040?

Transcript of opening address, ASCO Annual Meeting 2040, Chicago.

The speaker at the podium gestures at the screen behind them.

“I want to read three numbers into the record. Ninety-one: the percentage of patients across the top fifteen solid tumors who are alive at five years. Seventy-eight: the same number for pancreatic adenocarcinoma, the disease that, when many of you in this room entered fellowship, had a five-year survival in the single digits. And two-point-one: the average number of Stage IV solid-tumor presentations logged by a Memorial Sloan Kettering fellow last year compared to eighteen in 1996.

“That third number is supposed to alarm us. Some of our fellows have never seen tumors in the wild, only trained against them in simulations. And to that I say good.

“When I was a fellow at Dana-Farber in 1998, I rotated through a Stage IV pancreatic ward of sixteen beds. I learned more about dying than about cancer. I was trained by Dr. Robert Hsieh; at his retirement, he said something I never stopped thinking about.”

“He said: ‘If you do this job well, you will put yourself out of business. And I hope to God you do.’”

“A third-year surprised me at last night’s welcome reception. She said, ‘I’m going to Lagos to study history.’ When I asked her what she meant, she told me she wanted to see the things we have eradicated here, which meant going to the remaining places where AI screening is still not ubiquitous. Living in the future means having to travel to see the past.”

“AI-augmented multi-cancer detection has been deployed at population scale for just a bit more than half of a decade. It has changed so much of human life that historians of medicine are already trying to figure out what to call it. When they realize we already have a name we call it in private and ask us what we call it, we should tell them: ‘The Detection Decade’. No, not the decade where we finally cured cancer. The decade we caught it first.”

Subscribe now

© 2026 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.