@edzitron.com is on a roll! www.wheresyoured.at/premium-ai-i...

"Cognitive Surrender" trial: "Overall, across 1,372 participants and over 9,500 individual trials, the researchers found subjects were willing to accept faulty AI reasoning a whopping 73.2 percent of the time, while only overruling it 19.7 percent of the time." arstechnica.com/ai/2026/04/r...

I hope Adam McKay is preparing a lovely warm bath for Ed when he has to explain this to a general audience in a few years time.

Brilliant article, thank you!

Wow! What a great article. I wish every journalist would read it and then introspect deeply! Thank you! www.thehandbasket.co/p/refusing-t...

Another one of those what-if-it-happened-in-another-country moments: The president is suing the government for $100 million dollars for searching his home for the classified documents he stole, and the attorney who represented him on those charges is currently leading the Justice Department.

Warning to open source maintainers: the Axios supply chain attack started with some very sophisticated social engineering targeted at one of their developers simonwillison.net/2026/Apr/3/s...

Using @surf.social, I was within about 2 minutes able to create a new feed of #Artemis2 posts, then publish that feed to Bluesky. This is really something. Maybe the first time I've experienced what the open social web really could be. bsky.app/profile/did:...

Can someone tell me why we’re giving NASA’s funding to white supremacist Elon Musk? His rockets blow up. NASA’s don’t. #Artemis

This is so, so well-articulated.

This is the key for me. "AI" enthusiasts will read this and say "see, even critics admit it's useful". That would be a profound misreading—the point is how powerful institutions encourage/compel people to compromise their professional and ethical judgment, and how we need to fight those collectively

New blog post: Never underestimate what people who pay their lawyers by the year can get away with. On Elon Musk and Legal Arbitrage davekarpf.beehiiv.com/p/on-elon-mu...

Excellent read, thank you @taggart-tech.com taggart-tech.com/reckoning/

I appreciated this article by Michael Taggart, on the temptation to use some of these coding tools and the pitfalls, from the POV of an experienced software engineer. This is why it's important to fight these issues collectively rather than at the individual level. taggart-tech.com/reckoning/

Guess we all knew Anthropic had no idea how their models work... but nice to see it! > This code is so fucking funny[.] I am laughing so hard at how much of them are like "don't break the law, please do not break the law, please please please be good!!!!" neuromatch.social/@jonny/11632...

fly me to the moon but don’t make me use Outlook (new) I’d rather just use Discord to talk to the ground crew www.404media.co/artemis-2-as...

The. Rise. Of. Cognitive. Surrender. Study finds that people who use GenAI chatbots rely on them 80% of the time, and develop almost no capacity to recognize when a chatbot is feeding them faulty information. papers.ssrn.com/sol3/papers....

(One of) the dangers of using current AI's Thanks @garymarcus.bsky.social for finding this paper on cognitive surrender papers.ssrn.com/sol3/papers....

AI Insiders Are Preparing For The Bubble To Burst open.substack.com/pub/planetea...

It also raises the question of whether the guardian angel agents will need super guardian angel agents, that then are monitored by super-duper guardian angel agents, which are .... 🤦‍♂️🤦‍♂️

So they tried to use AI for strategy but it just didn't work due to the ubiquitous limitations of #GenAI Thanks @garymarcus.bsky.social , I'd venture that it was probably the combination of all 3 factors you mention! open.substack.com/pub/garymarc...

Great strategy, @microsoft.com ! You will surely win the AI race! 🤪🤦🏻‍♂️

Free newsletter: The Subprime AI Crisis is here. I explore the disturbing parallels between the run-up to the global financial crisis and the heavily-subsidized AI bubble, and how Anthropic and OpenAI's rush to IPO accelerates the crisis further. www.wheresyoured.at/the-subprime-ai-crisis-is-here/

The Subprime AI Crisis is here www.wheresyoured.at/p/7f2459f9-1...

Das ZDF verlangt von Mitwirkenden, „weder direkt noch indirekt mit Personen zusammenzuarbeiten, die auf einer (...) internationalen Sanktionsliste stehen“. Das ist nicht nur wegen der Art, wie Trump mit solchen Listen Politik macht, problematisch. www.sueddeutsche.de/medien/mitwi... (€)

Over at the newsletter I wrote about the unholy scourge that is "CEO said a thing!" journalism:

Wow! Every journalist needs to read this and question themselves — sadly, most won’t karlbode.com/ceo-said-a-t...

Thanks to @garymarcus.bsky.social for directing attention to Karl Bode's critique (@karlbode.com): ""CEO said a thing!" journalism involves parroting the claims of a business leader or executive with absolutely no context, correction, or challenge whatsoever..." karlbode.com/ceo-said-a-t...

"It seems to me that if one wants to build artificial general intelligence, it might help to know what intelligence is" True words!! Blind Squirrel Seeks a Theory of Intelligence open.substack.com/pub/herbertr...

Please tell me you’re cooking the books without telling me you’re cooking the books Goes for both @anthropic.com and OpenAI — plus prob many (all?) other AI companies Thanks @edzitron.com

If you think AI is stupid bullshit and you want to hear someone describe - in meticulously researched detail - exactly why it is stupid bullshit, have I got a podcast recommendation for you.

I’m sure Disney is happy! 🤦🏻‍♂️

Oh look I was right. Oh well, just $1 billion that never gets wired to anybody after an endless media hype cycle about the death of hollywood. Pobody's Nerfect! I'm sure everybody who wrote scary stories about Sora 2 destroying Hollywood will apologize www.hollywoodreporter.com/business/dig...

This will end up being likely a monologue and part of the intro of this week's premium, but every single person that wrote a "Sora 2 is destroying Hollywood" piece should be apologizing to their audiences and to the entertainment industry for deliberately misleading and scaring people in TV/Cinema

obligatory

The Future of Life Institute has completely rebranded and infiltrated journalism & labor spaces like there's no tomorrow. "one of the groups mentioned in the story, the Future of Life Institute, which is dedicated to warning about AI risks." 🙄 www.semafor.com/article/12/0...

the rebranding of all these organizations -- along with a rewriting of their history -- is so frustrating to see. ESPECIALLY w/ future of life institute & singularity institute (which is now known as MIRI but they try not to talk about the singularity part anymore ever since it went out of vogue)

"The AI Grad Student": A Harvard professor describes working with Claude. Early on, he describes misconduct that would cause any student to be terminated: "It faked results, hoping I wouldn't notice." But he ends the essay with "Now I'm doing 100% of my research with LLMs". Am I losing my mind?

I guess Disney amongst others will be happy about this …

Dear Bernie Sanders, Neil deGrasse Tyson, and Anti-AI Protestors: Please Stop Siding with PRO-EXTINCTIONISTS. An Open Letter. open.substack.com/pub/xriskolo...

Dear Bernie Sanders, Neil deGrasse Tyson, and Anti-AI Protestors: Please Stop Siding with PRO-EXTINCTIONISTS. An Open Letter. open.substack.com/pub/xriskolo...

There are a tiny handful of top editors and executives in the news business who've truly met this moment. @katie-drummond.bsky.social is one of them. Proud to play a small part on team @wired.com www.nytimes.com/2026/03/17/b...

Free newsletter: Why're we still doing this? Despite $1tr+ in investment, every AI company is unprofitable, LLMs have yet to provide tangible productivity benefits, and private credit-backed data center debt may be the next great financial crisis. www.wheresyoured.at/why-are-we-still-doing-this/

"Katie Drummond is leading Wired magazine through its best era." - @davekarpf.bsky.social davekarpf.beehiiv.com/p/on-the-wir...

not invented yet, and unlikely it'll ever be invented

One other note from today's newsletter: I'm worried about over-use of LLMs in software engineering. If people aren't reading (or are barely reading) code they ship, "writing code fast" becomes a liability. It also can stop you from learning to solve problems. www.wheresyoured.at/why-are-we-s...

@edzitron.com asks a very pertinent question: Why Are We Still Doing This? Pure madness! www.wheresyoured.at/why-are-we-s...