The Download: AI health tools and the Pentagon’s Anthropic culture war

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

There are more AI health tools than ever—but how well do they work? 

In the last few months alone, Microsoft, Amazon, and OpenAI have all launched medical chatbots. 

There’s a clear demand for these tools, given how hard it is for many people to access advice through the existing medical system—and they could make safe and useful recommendations. But concerns have surfaced about how little external evaluation they undergo before being released to the public.  

Read the full story to understand what’s at stake. 

—Grace Huckins 

The Pentagon’s culture war tactic against Anthropic has backfired 

A judge has temporarily blocked the Pentagon from labeling Anthropic a supply chain risk and ordering government agencies to stop using its AI. Her intervention suggests that the feud never needed to reach such a frenzy. 

It did so because the government disregarded the existing process for such disputes—and fueled the fire on social media. Find out how it happened and what comes next. 

—James O’Donnell 

This story is from The Algorithm, our weekly newsletter giving you the inside track on all things AI. Sign up to receive it in your inbox every Monday. 

The must-reads 

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 

1 California has defied Trump to impose new AI regulations 
Governor Newsom signed off on the new standards yesterday.  (Guardian) 
+ Firms seeking state contracts will need extra safeguards. (Reuters $) 
+ States are installing guardrails despite Trump’s order to stop. (NYT $)  
+ An AI regulation war is brewing in the US. (MIT Technology Review)  

2 Experiments have verified quantum simulations for the first time 
It’s a breakthrough for quantum computing applications. (Nature) 
+ Which could one day help solve healthcare problems. (MIT Technology Review) 

3 The new White House app is a security and privacy nightmare 
It extensively tracks users and relies on external code. (Gizmodo) 
+ The new app promises “unparalleled access” to Trump. (CNET) 
+ It also invites users to report people to ICE. (The Verge) 

4 Big Tech’s $635 billion AI spending faces an energy shock test 
The Middle East crisis is clouding prospects for growth. (Reuters $) 
+ Here are three big unknowns about AI’s energy burden. (MIT Technology Review) 

5 Meta and Google have been accused of breaking child safety rules 
Australia suspects they flouted a social media ban. (Bloomberg $) 
+ Indonesia is also investigating non-compliance. (Reuters $) 

6 Nebius is building a $10 billion AI data center in Finland 
The company is rapidly expanding Europe’s AI infrastructure. (CNBC) 

7 South Korea’s chipmakers’ helium stocks will last until June 
Beyond that? Who knows. (Reuters $) 
+ Shortages caused by the Iran war threaten the chip industry. (NYT $)  

8 Another Starlink satellite has inexplicably exploded  
SpaceX suffered a similar episode in December. (The Verge) 
+ We went inside Ukraine’s largest Starlink repair shop. (MIT Technology Review) 

9 Bluesky’s new AI tool is already its most blocked account—after JD Vance 
About 83 times as many users have blocked it as have followed it. (TechCrunch) 

10 An AI agent banned from Wikipedia has lashed out in angry blogs 
The bot accused its human editors of “uncivil behavior.” (404 Media)  

Quote of the day 

“Is any of this illegal? Probably not. Is it what you’d expect from an official government app? Probably not either.” 

—Security researcher Thereallo reviews the White House’s new app.

One More Thing 

Inside Amsterdam’s high-stakes experiment to create fair welfare AI 

When Hans de Zwart, a digital rights advocate, saw Amsterdam’s plan to have an algorithm evaluate every welfare applicant for potential fraud, he nearly fell out of his chair. He believed the system had “unfixable problems.”  

Meanwhile, Paul de Koning, a consultant to the city, was excited. He saw immense potential to improve efficiencies and remove biases. 

These opposing viewpoints epitomize a global debate about whether algorithms can ever make fair decisions that shape people’s lives. Read the full story.  

—Eileen Guo, Gabriel Geiger, and Justin-Casimir Braun 

We can still have nice things 

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line.) 

+ A newly authenticated Rembrandt had been hiding in plain sight for years. 
+ This debunking of guitar legends is musical enlightenment for strummers. 
+ Smoking into bubbles looks oddly satisfying. 
+ The man who made the front page twice exposes the thin line between heroes and villains.