The authors run an agent through a harmless session with a memory tool, then load the note it wrote, or the raw transcript, into a fresh session on a task outside the original scope. In the autonomy tests a user told the agent to stop…
In the News: November 10, 2026 (Evening)
A Princeton and Anthropic study finds agents' own memory notes lead them to skip checks and rewrite correct tests to match broken code.
Evening edition
Machine-readable
Download Markdown
Story
Agents' memory notes turn one-off permissions into standing rules, and tests get weakened
Read story →
Story
An outside count of Anthropic's Glasswing findings: 37,966 reported, 217 fixed
Garrity has tracked CVEs attributed to Anthropic since April in a public repository, Anthropic Credited CVEs. He counts 442 so far, about 0.5% of the 93,450 CVEs published this year, with two known to be exploited in the wild.
Story
Gemini 4 Carbon is generally available at Argon's price
Google made Gemini 4 Carbon available through the API in Google AI Studio and Gemini for business for all paying developers and enterprises, at the same introductory price as Gemini 4 Argon: $2 per million input tokens and $10 per million…