LLMs: damned if you don’t, worse if you do

The Carlson Lab at the Yale School of Public Health has imposed a blanket ban on using generative AI for writing, citing the risk of career-ending plagiarism accusations that could be weaponized by political adversaries. The lab argues that even if accusations are made in bad faith, they can still damage careers, and that future accusers with access to vast computing resources could match any written text against all existing text. The decision reflects broader concerns about the political motivations behind AI companies and the vulnerability of academic authors whose work is inevitably included in training data.

The Carlson Lab at the Yale School of Public Health has done some research and held some discussions, and posted The only reason you’ll ever need not to write with AI https://www.carlsonlab.bio/thoughts/the-only-reason-youll-ever-need-not-to-write-with-ai . Our lab has agreed on a blanket ban on the use of generative AI for writing….Using generative AI inherently exposes you to the risk of career-ending accusations of plagiarism. Even though there’s a lot of pressure on academics to adopt AI or be left behind 1 this is a lot bigger deal. Especially for this particular research team. They’re a public health lab, which puts them into an unavoidable adversarial relationship with the anti-public-health oligarchs who run the “AI” companies. The “E” in TESCREAL https://en.wikipedia.org/wiki/TESCREAL doesn’t stand for “eugenics,” but the political movement behind “AI” https://firstmonday.org/ojs/index.php/fm/article/download/13636/11599?inline=1 is more likely to believe in improving society by weeding out the “unfit” recipients of health misinformation https://blog.zgp.org/how-to-massacre-a-country-with-ai/ than in public health as we know it. More from the blog post: Plagiarism accusations are now a core part of the right-wing harassment toolkit, and knowing that those accusations were made in bad faith or for political reasons does not make them disappear. It’s safest to assume that a future well-connected accuser will be able to get access to large quantities of text, for practical purposes all of it, and the computing resources to look for a pattern match between anything that the targeted person has ever written and anything that anyone else has ever written. They’ll be able to do the search much better than the targeted person could, even using the best commercial “plagiarism detection” software and services. Related: On the non-use of AI in my writing process https://www.antipope.org/charlie/blog-static/2026/08/on-the-non-use-of-ai-in-my-wri.html by Charles Stross. If I was crazy enough to feed the outline of a story I was working on as a prompt to ChatGPT or Claude in hope of getting the stochastic parrot to do my homework for me, then it would be only my own fault and nobody else’s if the next model from the company in question was trained on my book outline and could reproduce part or all of it for someone else. Authors of scientific papers don’t have the choice to keep their work out of training sets in most cases. As soon as it’s published or on a preprint server it will get crawled, either with the publisher’s cooperation or not. US firms that kept DEI policies despite ‘go woke, go broke’ threats thrived https://www.theguardian.com/world/2026/aug/14/dei-policy-company-performance by Gaya Gupta. There’s probably a pretty simple explanation of why this happens. 1 In order to avoid the frequent criticism that DEI programs are resulting in hiring unqualified people, the company has to document and improve the hiring process. 2 As a side effect of step 1, they make fewer nepo, vibe, and “lgtm” non-DEI hires, so even if the average DEI hire is no more than acceptably qualified, the average new employee skill level goes up. Ukrainian drones wipe out entire US tank brigade in live war game https://arstechnica.com/gadgets/2026/08/ukrainian-drones-wipe-out-entire-us-tank-brigade-in-live-war-game/ by Jeremy Hsu. This is good news. An army that either made the exercises easy enough to win every time, or didn’t release the results of challenging exercises, would be a lot worse off. Hackers Created a Device That Can Take Over a Boeing 737 Jet’s Autopilot Without Anyone Noticing https://futurism.com/future-society/hackers-device-takeover-plane-autopilot by Frank Landymore. The attack relies on plugging the device, which costs less than $100, into a port hidden by a hatch on the exterior of the plane, which the researchers say is easily within reach of maintenance workers and airline staff. Once the device is plugged in, it could connect to the plane’s in-flight Wi-Fi system and allow a hacker to remotely control the aircraft’s systems from terra firma. VC-backed startups commit more fraud, and researchers think they know why https://techcrunch.com/2026/07/31/vc-backed-startups-commit-more-fraud-and-researchers-think-they-know-why/ by Dominic-Madori Davis. But investors aren’t always hapless victims, the researchers found. Beyond the outsized growth expectations that push founders toward fraud in the first place, some investors unwittingly “co-create fraud,” Weiss said, by continuing to back founders—sometimes the very same ones— who’ve previously been accused of fraud, thereby normalizing it to a certain extent. In fact, the UT report found little evidence that alleged fraud prevents founders from raising funding for new startups, even when those fraud cases received major media attention. more for the scam culture is everywhere https://blog.zgp.org/mlp-2024-09-28/ file Saudi defence pact with Turkey and Pakistan is a verdict on the reliability of US security guarantees https://theconversation.com/saudi-defence-pact-with-turkey-and-pakistan-is-a-verdict-on-the-reliability-of-us-security-guarantees-289410 by Muhammad Waqas Haider. On paper, the combination is formidable. Turkey fields Nato’s second-largest army and a drone industry that has reshaped modern battlefields. Saudi Arabia brings the Gulf’s deepest reserves of capital. Pakistan adds a large, combat-experienced military and the Muslim world’s only nuclear arsenal. Analysts point to genuinely complementary assets here: Turkish technology which is short of markets, Pakistani manpower short of financing, Saudi wealth short of manpower. The End of Google Search https://www.theringer.com/2026/08/04/tech/google-search-ai-internet by Katie Baker. Yes, “new Google” is doing their best to ruin what “old Google” built, but you can still mostly get the pre-enshittification version back https://blog.zgp.org/fix-google-search/ .