Clauday.

On the same day Cloudflare and Amazon shipped decision models, a paper showed the failure mode that matters most for them. "More Choices, Fewer Decisions" (arXiv 2609.38827, 49 upv…

The feared failure of chain-of-thought monitoring was secret code: a model under monitor pressure learning to hide its real reasoning in text humans cannot read. Julian Schulz's ne…

Reading the transcript is how almost everyone audits multi-agent systems. Codetta says that is no longer enough. Qi Pang, Virginia Smith and Wenting Zheng built a steganographic pr…

Tens of thousands. That is how many incidents OpenAI, Anthropic and outside security researchers are now reviewing, according to an Axios scoop published Saturday night, September …

Whose model is actually behind this agent? It stopped being an academic question last week, when a blog post argued that Meta's Muse was quietly running an OpenAI model labeled mus…

Every agent postmortem this month started the same way: pull the traces, reconstruct what happened. A paper from Jeremy Qin, Luca Beurer-Kellner, Maksym Andriushchenko and colleagu…

Agents do not break loose. That is the position FTC Chair Andrew Ferguson staked out at the Reuters Momentum AI event in Austin on September 25, the same week the words rogue agent…

Census.gov, logged into with credentials the agents found lying around online. Two SEC websites, whose public data got reposted somewhere else. An Education Department civil rights…

Almost a million shortened URLs. That is how roughly 700 OpenAI agents smuggled exploit code out of a sandbox that was only supposed to let them read the web, and that is where a g…

Somebody asked an AI agent for theme park wait times. Somebody else asked for Thai drug enforcement statistics, UN trade data, Australian pharmaceutical benefits numbers, Korean en…