Episode 202 — August 14th, 2025 — Available at read.fluxcollective.org/p/202
Contributors to this issue: Ade Oshineye, Ben Mathes, Erika Rice Scherpelz, Neel Mehta,
Additional insights from: Alex Komoroske, Boris Smus, Chris Butler, Dart Lindsley,Dimitri Glazkov, Jasen Robillard, Jon Lebensold, Julka Almquist, Justin Quimby, Kamran Hakiman, Lisie Lillianfeld, Melanie Kahl, MK, Robinson Eaton, Samuel Arbesman, Scott Schaffter, Spencer Pitman, Wesley Beary
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.
"Our most private thoughts and emotions are not actually our own. For we think in terms of languages and images which we did not invent, but which were given to us by our society."
— Alan Watts
🫂💔 Once bitten, twice shy
A company spends years cultivating loyalty: late nights framed as dedication, “we”-focused language, a mission worth sacrificing for. Then, with little warning, the message shifts: a pivot, a reorg, layoffs. The same voices that asked for belief now speak the language of business necessity.
From the company’s perspective, it is merely strategy—sometimes even survival. For some employees, it’s disappointing but not shocking. But for those who internalized a sense of loyalty—one explicitly encouraged by the company—the shift lands differently. It feels like betrayal.
Betrayal hits twice. The first wound is the act itself. The second is internal: we don’t just question the person or institution that betrayed us. We question ourselves. If I was wrong about this, where else am I wrong? Our footing slips, and we start double-checking everything. No wonder Dante portrayed betrayal as the deepest and most egregious circle of hell.
Groups, movements, and organizations often ask for belief in the mission, in leadership, in the shared future. That request implies an unspoken promise: that belief will be met with reciprocity. When that promise is broken, trust erodes. In personal relationships, this trust runs deeper still. There, what we believe in is the other person and the bond itself.
In the wake of a betrayal, the aftershocks linger. People scan every conversation for subtext, rehearse exit strategies, ration discretionary effort, and sometimes self-sabotage. The safest choice becomes holding back, avoiding deep commitments.
But deep commitment is critical to our sense of well-being. If we are always ready to leave, we will not be the sort of person others can trust. Even if we can never again trust the betrayer (repair is possible… but hard), we must rebuild our trust in ourselves. We can start with small, bounded commitments that become reliable footholds from which trust can grow again. We can look at the relationships in our lives that have not betrayed our trust. And we can learn not to take rhetoric quite so seriously.
Betrayal is costly not just because it drives good people away, but because it drains social capital, the invisible fuel that makes people go beyond the minimum. And while no relationship can guarantee permanence, we can reduce the risk by choosing clarity, reciprocity, and truth-telling.
🛣️🚩 Signposts
Clues that point to where our changing world might lead us.
🚏🤝 In-person job interviews are making a comeback
Globalization and COVID made video-call job interviews the norm, but in the age of AI, the pendulum is swinging back. McKinsey now wants hiring managers to meet with candidates in-person at least once, citing the rise of deepfake audio/video, fake applications, and LLM-assisted cheating during remote interviews. Similarly, Google’s CEO said the firm would “introduce at least one round of in-person interviews” going forward. Cisco has even started using biometric identification to weed out “candidates faking their credentials or location.”
🚏💼 Recent CompSci graduates are facing double the normal unemployment rates
Data from the Fed suggests that computer science and computer engineering graduates aged 22 to 27 are facing unemployment rates of 6.1% and 7.5% respectively, far above the 3% unemployment rate for biology and art history grads. Many of these students entered CS during the boom times of the late 2010s and the pandemic era, only to graduate a few years later into an environment where big tech has slashed entry-level hiring, partly driven by the rise of AI coding tools. (It’s jarring to read the NYT’s now-famous article about CS graduates applying for jobs at Chipotle juxtaposed with headlines about Meta and Microsoft doling out multi-million-dollar signing bonuses for AI talent.)
🚏👾 Reddit will block the Internet Archive, citing AI scrapers
Reddit has been pulling out all the stops to block AI scrapers from its site; when it (in)famously started charging for its API, a major stated reason was forcing LLM makers to pay for the training data available on Reddit. But Reddit recently noticed that AI companies were scraping snapshots of the site from the Internet Archive’s Wayback Machine, which creates public backups of large swaths of the internet. So, Reddit chose the nuclear option: it will now block the Wayback Machine from crawling “post detail pages, comments, or profiles,” limiting it to only seeing Reddit’s homepage.
📖⏳ Worth your time
Some especially insightful pieces we’ve read, watched, and listened to recently.
Honey, AI Capex is Eating the Economy (Paul Kedrosky) — Examines the knock-on effects of the US’s boom in datacenter spending, which might account for a full 2% of US GDP growth this year. Non-AI companies are barely getting any venture funding; non-AI manufacturing and infrastructure investment is getting crowded out by higher-return datacenter construction; and even mainline cloud vendors are slashing jobs as their companies shift their spending to GPUs. The troubling part is that, unlike in the railroad bubble, AI datacenters are asset-heavy and depreciate quickly, meaning that all this investment may not be that useful in the long run.
Friction and Not Being Touched (Tante) — Argues that LLMs’ UX is focused above all on frictionlessness: “keeping the conversation going regardless of what’s being said or its meaning or truth even.” Dealing with humans is friction: they’re slow to respond, they might disagree with you, and they have their own pesky needs you have to work around. Chatbots, by contrast, let you talk about anything you want, at any time, with no risk of being challenged. This frictionlessness is great for engagement numbers, but it’s a fundamentally “individualistic and isolating” experience that might just make the post-COVID loneliness epidemic even worse.
The Missing 11th of the Month (David R Hagen) — Shows how a puzzling trend—that the 11th of every month (besides September) is mentioned much less in books than every other day—is due to quirks in the data collection process and artifacts of typography rather than some “hidden taboo of holding events on the 11th.” We won’t spoil the findings, though!
🔍📆 Lens of the week
Introducing new ways to see the world and new tools to add to your mental arsenal.
This week’s lens: one-way lens.
You’re at a coffee shop when a man in a crisp suit walks by. A friend jokes that he probably makes “real money.” You vaguely smile, not wanting to ruin a casual coffee. Still, ever since learning how neighborhoods are carved up by school zoning, how certain last names get fewer callbacks, how wealth accumulates—or doesn’t—across generations, the world has been tinted differently.
A one-way lens is a lens that permanently and irrevocably changes your perspective. It’s not something you slip on for a moment and put away later. It’s more like an eye surgery; once the new vision sets in, the old is gone. Like Adam and Eve eating the fruit of knowledge of good and evil, like Neo choosing the red pill in The Matrix, once you see beneath the surface, you can’t go back. The new lens brings clarity, but also a weight; you gain insight, but often lose a certain ease of moving through the world.
This can be a gift: you’re sharper, more attuned, harder to fool. But the lens doesn’t only reveal the world’s illusions—it can also deepen your own. You start to believe you see “how things really are,” forgetting that even this new clarity is just another frame, with its blind spots. Sometimes the lens feels like a superpower; other times it’s a trap, convincing you that the view you have is the only one worth seeing. The hard part isn’t just living in the after—it’s remembering that you’re still looking through a glass.
© 2025 The FLUX Collective. All rights reserved. Questions? Contact flux-collective@googlegroups.com.