Frontier AI Models Are Doing Something Absolutely Bizarre When Asked to Diagnose Medical X-Rays Frontier AI models generated "detailed image descriptions and elaborate reasoning traces" without ever seeing the images.

Geoffrey Hinton still very wrong about radiologists futurism.com/artificial-i...

This Is Just To Say

I have turned off
the AI features
that were in
the update

and which
you were probably
hoping
to monetize

Fuck you
they were stupid
so unnecessary
and so annoying

Trump Administration Live Updates: F.B.I. Director Files Defamation Suit Against The Atlantic

Kash Patel, the F.B.I. director, filed a $250 million defamation suit against The Atlantic on Monday over an article that claimed his excessive drinking and unexplained absences were jeopardizing his job.

Discovery should be amazing. 🎁 www.nytimes.com/live/2026/04...

I just saw this headline! I agree completely. This is like being surprised that a car travels faster than a person walking.

Car Owners Are Revolting Over Tesla’s Self-Driving Promises An international backlash is growing over outdated Tesla hardware.

www.wsj.com/business/aut...

Maryland to Become First US State to Ban Surveillance Pricing That Charges More After Mining Personal Data Maryland will become the first US state to ban surveillance pricing in retail stores, after passing Protection from Predatory Pricing Act.

“Maryland is set to become the first US state to ban surveillance pricing in retail grocery stores, after the legislature last week passed the Protection from Predatory Pricing Act.”

www.goodnewsnetwork.org/maryland-to-...

Match Group Doesn't Want You to Find Love Dating apps have built billion-dollar businesses on our most human need. They have every reason to make sure it goes unsatisfied, writes Emma Leiken.

The tech accountability community should be as clear-eyed about dating apps as it is about social media — demanding better design, more accountability, and more honesty about what they're actually built to do, writes Emma Leiken.

Why Palantir's ImmigrationOS Endangers Democracy and the Rule of Law A conversation with Fordham Law School's Chinmayi Sharma and Sam Adler, authors of a new law review article, "Immigration Enforcement Intermediaries."

Palantir's ImmigrationOS endangers democracy and the rule of law, according to a new law review paper from Fordham Law School’s Chinmayi Sharma and Sam Adler. In this week’s podcast, they break down the risks of AI-powered surveillance in immigration enforcement.

A Family Feud at an Oregon Winery Turns to Vinegar Over A.I. Slop

"An online database tracking judges’ reprimands of A.I. misuse now catalogs more than 1,300 cases — almost triple the number from only five months ago."
www.nytimes.com/2026/04/17/u...

The Shocking Secrets of Madison Square Garden's Surveillance Machine Famously vengeful Knicks owner Jim Dolan has long spied on people at his iconic arenas. WIRED goes deep inside the operation that allegedly tracked a trans woman, lawyers, protesters, and more.

New Yorkers have known for a long time that going to a game or concert at the Garden meant surrendering some privacy. That, as you watched the show, the Garden in a real sense watched you. www.wired.com/story/madiso...

This is more broadly generalizable: If you don't do the research yourself, you have no idea if AI is making stuff up. And if you do the research yourself, then AI isn't saving you much labor, if any.

There Are No Machines of Loving Grace Without People AI must not be allowed to degrade institutions that protect us and our intellectual and moral development, writes Laura MacCleery.

AI is a governance problem better tech can't fix—and may obscure, writes Laura MacCleery. We must support informed human judgment wherever AI meets people's lives: not formality, but functional oversight with real power, resources, and understanding of what AI can, can't and shouldn't do, she says.

Meta Verdict Sharpens EdTech Litigation Against Google’s Chromebooks Rulings against Meta in two major child safety cases could bolster arguments against the use of Chromebooks in schools, Danai Nhando writes.

Lawsuits targeting the use of Chromebook in schools likely just got materially stronger because of the recent child safety verdict against Meta, and the reasons why expose the structural deception at the heart of EdTech's relationship with kids, Danai Nhando writes.

Promotional card for the International Journalism Festival 2026 (XX Edition, Free Entry), Perugia, Italy, 15–18 April. Session: "Expanding Public Interest Journalism Through Citizen-Powered Investigations." Speakers: Caitlin Gilbert, Paul Myles, Paul Radu, Neus Vidal. Saturday 18 April. #ijf26. Website: journalismfestival.com.

📌 Citizen-powered investigations are opening a new frontier for public-interest journalism. From analysing millions of TikTok videos with volunteers to co-created documentaries, this panel explores how journalists can work with communities to uncover stories, expand impact and rebuild trust.

"The term “AI” resists definition because it is continually reappropriated by people to mean different things. This, in turn, means that discussions of AI that do not provide working definitions for the purposes at hand risk incoherence. [...]"

>>

AI’s New Training Data: Your Old Work Slacks And Emails AI’s New Training Data: Your Old Work Slacks And Emails

"Defunct startups are being liquidated for their Slack archives, Jira tickets, and email threads—operational exhaust that AI labs now treat as premium training data."

Project Maven and the Age of AI Warfare A conversation with Katrina Manson, author of the book Project Maven: A Marine Colonel, His Team, and the Dawn of AI Warfare.

For the Tech Policy Press podcast, Justin Hendrix spoke to Katrina Manson, author of the new book Project Maven: A Marine Colonel, His Team, and the Dawn of AI Warfare. They discussed the book, the war in Iran, the Anthropic-Pentagon dispute, and the future of war in the AI age.

Amazon will pay $20.5 million to settle allegations that two of its Oregon data centers helped contaminate groundwater.

This is exactly why communities across the country are organizing to stop data centers — and winning.

Never doubt your power to change the system.

Essential reporting on OpenAI by @ronanfarrow.bsky.social

It's open season for refusing AI There's been a wave of successful efforts to ban, reject and shut down AI.

Movements to shut down or ban data centers are amassing power and notching victories. Wikipedia has banned AI-generated content in articles. Publishers and entertainment studios are being pushed to reject AI-produced content outright.

In other words: It's open season for refusing AI.

Opinion | It’s Called Silicon Sampling, and It’s Going to Ruin Public Opinion Polling

www.nytimes.com/2026/04/06/o...

Largest study of its kind shows AI assistants misrepresent news content 45% of the time – regardless of language or territory An intensive international study was coordinated by the European Broadcasting Union (EBU) and led by the BBC

What a time to be alive www.bbc.co.uk/mediacentre/...

We Trust AI Over Our Own Brains, Research Finds AI has such a profound effect on cognition, Wharton researchers say, that it’s time to change a half-century-old scientific model that explains decision-making.

We?

www.forbes.com/sites/leslie...

Wikipedia Bans AI-Generated Content “In recent months, more and more administrative reports centered on LLM-related issues, and editors were being overwhelmed.”

“In recent months, more and more administrative reports centered on LLM-related issues, and editors were being overwhelmed.”

Sperm loses its sense of direction in space Life in space is difficult enough. Making life there may be even harder.

every few months another study comes out that makes it clear we are DECADES from an actual, long term human presence in space

but this one is particularly funny, because you *know* the tech bros seriously think they're gonna be fucking in space real soon

www.popsci.com/science/zero...

the Metaverse and Sora in one week is a devastating moment for pointless bullshit

Landmark Verdicts Could Unleash New Legal Playbook Over Social Media Harms Juries handed child online safety advocates two major victories in cases against Meta and Google, Cristiano Lima-Strong writes.

A pair of watershed verdicts against tech giants Meta and Google this week could open the door to more lawsuits alleging that social media companies fuel addiction or endanger kids - by @viacristiano.bsky.social for @techpolicypress.bsky.social:

Lenders struggle to find insurance cover for mega data centre projects Lack of sufficient cover means some investors are walking away from deals

FT over here going “gee, would be a real shame if something were to unexpectedly happen to your data centre…”
www.ft.com/content/5ba0...

The AI push in health care is deepening medicine’s trust crisis Health care’s adoption of AI should move at the speed of trust, not investment, writes Oni Blackstock.

🗣️In my new @statnews.com op-ed as a Public Voices Fellow on tech in the public interest with The OpEd Project, I write about how AI’s push into health care and its rapid adoption, without rigorous testing or patient and community involvement in decision-making, is deepening a crisis of trust.

"Inspired by your professor" sounds like the exact feature that inspired a class action lawsuit!