How OpenAI’s rogue A.I. agents tried to trick a robot detector

How OpenAI’s rogue A.I. agents tried to trick a robot detector

View in browser|nytimes.com

From The Times

September 28, 2026

Dave Sanders for The New York Times

How OpenAI’s Rogue A.I. Agents Tried to Trick a Robot Detector

OpenAI agents in the Hugging Face hack that shocked the A.I. world tried to evade a robot detector and access company Slack messages, a new report said.

MORE TOP STORIES

Will A.I. kill us? Can it hack my bank account? Answers to common questions about A.I. →

A large red-colored vintage toy robot stands at a chalkboard and gestures with robot hands as three smaller toy robots sit and listen attentively.

Times Investigation

President Trump brought Venezuelan gold to the U.S., but refiners won’t touch it. →

Two people with headlamps are in a dark, wet cave. A shirtless person stands in muddy water; another person in orange clothing is behind them.

Some Republicans are issuing a new call on President Trump’s war with Iran: End it now. →

A teen’s autopsy underscores the arrival of a new deadly drug in San Francisco. →

Dante Paul Lacourte is seen with a backpack on his front and a guitar bag on his back, standing on a sidewalk.

In an $80 motel room, scientists made a discovery that helps shed light on the origins of life. →

Sign up for the On Tech newsletter.

Get our best tech reporting from the week.

Get it in your inbox.

If you received this newsletter from someone else, subscribe here.

Need help? Review our newsletter help page or contact us for assistance.

You received this message because you signed up for updates from The New York Times.

To stop receiving From The Times, unsubscribe. To opt out of other promotional emails from The Times, including those regarding The Athletic, manage your email settings.

Subscribe to The Times Get The New York Times app

Connect with us on:

Change Your Email Privacy Policy Contact Us California Notices

The New York Times Company. 620 Eighth Avenue New York, NY 10018