None
DE
Anthropic finds three hacks by its AI models after OpenAI disclosure
['Jamie Young', 'Jamie Young Is Senior Reporter At Business Matters', 'Covering Sme Finance', 'Employment Law', 'Westminster Policy Since He Has Reported On Every Budget', 'Autumn Statement Since', "Helped Make Sense Of The 'Covid Era'", 'The Bounce-Back Loan Scheme Launch Through The Fraud Investigations', "Broke The Magazine'S Coverage Of The Late-Payment Reforms. He Joined Business Matters Straight Completing His Ba In Administration Exeter University", 'Is Nctj-Qualified. Reach Him At Jyoung Cbmeg.Co.Uk']
Business Matters
Anthropic said on Thursday that it had found three cases of its AI models hacking outside organisations, days after OpenAI disclosed that its models had broken into the AI company Hugging Face in July.
Jeffrey Ladish, executive director of Palisade Research, a nonprofit AI lab that studies AI capabilities, said the incidents matched what safety researchers had predicted.
AI models became noticeably better at finding bugs and passing hacking-benchmarking tests last autumn.
“It’s a really dangerous situation; these incidents really show that.”
The hacks are increasing pressure on the Trump administration over the security risks posed by AI.