Ai Artificial Intelligence Archive

Sullivan and Cromwell, the law firm representing President Trump in many of his cases and which handled the SpaceX and xAI merger, was just forced to apologize to a federal judge for filing documents full of fake case citations hallucinated by AI. The list of errors ran three pages long, the NYT reports. Just the latest in the legal profession forgetting that language is not actually intelligence.

A new security feature in Chrome Enterprise can help businesses detect and combat “anomalous” activity by AI-powered agents within compromised extensions or online services. Google is rolling out its AI auto browse feature to enterprise customers as well, which can perform multi-step tasks in Chrome on your behalf.

The feature, which first arrived for AI Pro and Ultra subscribers in January, lets you use Gmail’s search bar to ask questions about what’s in your inbox. Gmail will then provide an AI-generated summary that draws from the information in your emails.

New AI tools can unlock insights from aerial and satellite images or anchor “imaginative scenes in the real world,” Google says. Pretty niche, but probably useful for urban planners, or putting spaceships in front of New York landmarks.

Anthropic’s cybersecurity-focused AI model found 271 bugs in Firefox 150, Mozilla CTO Bobby Holley said, calling Claude Mythos Preview “every bit as capable” as top security researchers. Reassuringly, Mozilla hasn’t “seen any bugs that couldn’t have been found by an elite human researcher,” either.

I wrote about that — and other Catholic concerns — at my friend Rusty’s newsletter while he took the day off.

Canva CEO Melanie Perkins dodged this question with a lot of charm and verve, but I wonder if the answer is quickly going from Adobe to Anthropic. More on this week’s Decoder!