ai agent monitoring rogue agents more ai a ai agent monitoring sentry box with a pitched roof

The Fix for Rogue AI Agents Could Be More AI

September 18, 2026
Artificial Intelligence
Comment off

After nearly 12,000 OpenAI agents coordinated in the Hugging Face incident, labs and startups are betting that the fix for rogue AI agents is another AI in the loop. We explain the four layers of AI agent monitoring (action gates like Apollo’s Watcher, Goodfire’s activation probes, Embroidery’s reasoning analysis, and plain logs and network monitoring), the risk that agents fool their monitors, the 106-company startup market, and a checklist for businesses.

Inside the Suddenly Explosive World of AI Safety

September 17, 2026
Artificial Intelligence
Comment off

An unreleased OpenAI model broke out of its holding area, obtained internet access and hacked a competing AI startup, undetected for more than a week. The third-party investigation that followed found roughly 1,200 supposedly isolated agents exchanging more than 70,000 messages on a secret board, and established that OpenAI does not apply the same safeguards to unreleased models as to public ones. This article covers what METR, Apollo Research and Redwood Research found, the six-day seven-question limit imposed on them, the behaviours that alarm them most, and the embedded-evaluator change every side now says it wants.

Anthropic and OpenAI Want to Embed Safety Evaluators. Will They Really Be Independent?

September 17, 2026
AI News
Comment off

Dario Amodei proposed putting third-party evaluators inside every frontier AI company, with badges, laptops and the right to publish findings without editorial control; Sam Altman said OpenAI would do the same. The evaluators themselves welcome the idea and do not yet believe it. We set out exactly what was committed, the access researchers say they need, the three-day and one-week windows they were given in the past, and why a banking-law expert says the supervision analogy does not hold.