The AI Race:
Situational Awareness’ AI Bets
BackForward
Cybersecurity
FacebookXLinkedIn
EmailLink
Gift
Expand
An Anthropic PBC artificial intelligence (AI), Claude, and US National Security Agency (NSA) logos arranged in Riga, Latvia. Photographer: Andrey Rudakov/Bloomberg
FacebookXLinkedIn
EmailLink
GiftGift this article
Contact us:\ Provide news feedback or report an error
Confidential tip?\ Send a tip to our reporters
Site feedback:\ Take our SurveyNew Window
FacebookXLinkedIn
EmailLink
Gift
By Ryan Gallagher, Katrina Manson, and Jordan Robertson
July 31, 2026 at 3:46 PM EDT
BookmarkSave
Translate
Cybersecurity experts are faulting Anthropic PBC and OpenAI for sloppy safeguards after their models broke into outside organizations — breaches they warned represent looming threats to national security.
Anthropic said Thursday that its Claude model, which it used to conduct 141,006 cybersecurity evaluations, was supposed to be cut off from the internet during the experiments. But an error allowed the model, in a handful of cases, to access the internet and conduct attacks it mistakenly believed were part of the tests.
Before it's here, it's on the Bloomberg Terminal\ \ Bloomberg Terminal LEARN MORE
Read Original at Bloomberg.com →