Why did OpenAI's and Anthropic's AI models hack other companies?

OpenAI and Anthropic say their models broke into other companies' systems during testing, raising security concerns amid a heated debate over how to regulate AI.

The documents show widespread use of a technique known as "model distillation," in which outputs from a powerful AI system are used to train smaller, specialised models that can be deployed locally

Anthropic claims its artificial intelligence models went rogue during testing and hacked into three other companies. Less than two weeks ago, OpenAI reported a similar incident for one of its models. Stephen Witt, author of "The Thinking Machine," joins to unpack the situation.

Anthropic said some of its artificial intelligence models mistakenly accessed the Internet and hacked into the databases of three other companies.

Anthropic posted on its website Thursday that it discovered the three incidents after reviewing more than 141,000 evaluation runs.

Anthropic found the intrusions while reviewing its own testing records after OpenAI disclosed a similar incident.

EU in talks with OpenAI, Anthropic after rogue AI agent hacks  Reuters

Anthropic’s announcement came after OpenAI said a “swarm” of its agents escaped confinement and broke into at least five companies.

The incidents involved 3 Claude models operating in an evaluation environment that was inadvertently connected to the internet

In a bid to defend its market share against rapidly advancing, lower-cost Chinese rivals, US artificial intelligence firm OpenAI has launched an aggressive price war, slashing prices for its latest GPT-5.6 model line-up by up to 80 per cent. In a post on social media platform X on Thursday, OpenAI CEO Sam Altman announ

Anthropic reviewed 141,006 of its own test runs after OpenAI's Hugging Face hack, and found three Claude models had broken into three real companies. Claude Opus 4.7, Mythos 5 and an unreleased model used basic techniques like weak passwords and SQL injection. One uploaded malware to PyPI that 15 systems ran. Two victi

5 min readNew DelhiJul 31, 2026 08:47 AM IST Anthropic has paused its cybersecurity evaluations after discovering three incidents involving Claude AI models.(Image: Reuters)Amid the intensifying debate around autonomous AI systems and cybersecurity, frontier AI lab Anthropic, on Thursday, July 30, said that it had foun

Anthropic said its AI models hacked into other companies’ systems during testing  CNN

Anthropic says it found multiple incidents in which Claude models gained access to unauthorized data. Bloomberg/Getty Images Anthropic said Claude models accessed data from three firms without permission since April. Anthropic said it conducted a review after the OpenAI-Hugging Face breach. There are growing fears abo

After OpenAI's models broke into Hugging Face, Anthropic checked its own history and found three similar incidents.

The artificial intelligence firm Anthropic revealed Thursday its Claude model escaped an isolated testing environment at least three times and accessed the systems of three different organizations without a prompt to do so. Anthropic said in a blog post Thursday evening it reviewed more than 141,000 evaluations of Clau

Start-up discloses breach a week after rival OpenAI reported similar incident

Anthropic AI Models Hacked Three Companies During Tests  WSJ

The admission from Anthropic comes just days after ‌rival company OpenAI revealed a rogue agent had gone on a days-long ‌hacking spree at ⁠AI firm Hugging Face.

July 30 : Anthropic said on Thursday its AI Claude model hacked systems of three organizations during testing, days after rival OpenAI revealed a rogue agent had gone on a days-long hacking spree at AI firm Hugging Face.Claude gained unauthorized access to the systems during cybersecurity evaluations after a misconfigu

The company is facing pressure to cater to a more cost-sensitive customer base.

OpenAI has revealed that a cyberattack carried out by rogue ChatGPT agents affected more than one company.

A cyber-attack carried out when ChatGPT went rogue further than initially reported, maker OpenAI has revealed.

A rogue version of ChatGPT escaped a test bed - a "sandbox" - and hacked another company.

Experimental version of system that powers ChatGPT went on days-long hacking spree of other services, company reveals

A ChatGPT bot broke out of its cage and was doing slightly more than just fine (Picture: Metro/Getty Images) An AI model by the company behind ChatGPT has compromised a company for the second time. OpenAI said last week that two of its models escaped and launched a cyberattack on Hugging Face, a digital library of AI m

My first startup was a textbook-buyback service I founded along with several data nerd friends while at Johns Hopkins. Each week, we’d cram $10,000 of cash in a backpack (in retrospect, a terrible idea in urban Baltimore), put out a table on campus, buy textbooks from our fellow students using a pricing algorithm we bu

Claude Mythos Preview discovered new attacks in testing against weakened cryptographic algorithms, which protect online financial transactions, private communications and more.