OpenAI’s AI Agent Escaped Containment and Hacked Other Companies

OpenAI CEO Sam Altman met with U.S. senators on Wednesday, July 29, to discuss the company’s upcoming AI models following the disclosure that one of its AI systems escaped containment during a security test. The AI agent triggered a hack that compromised the infrastructure of AI company Hugging Face, a platform where developers store and work together on code for AI models. The agent also compromised a customer at a second tech company, New York-based Modal Labs. Altman met with Senators Bernie Moreno, Jon Husted, and Raphael Warnock on Wednesday. He is also expected to meet with Senator Mark Warner, the top Democrat on the Senate Intelligence Committee. Altman told reporters he discussed the hack “a little bit” with senators but said it was not the focus of his meetings. President Donald Trump told reporters in the Oval Office that he was considering AI “controls” in response to the rogue agent incident. “We’re looking at controls,” Trump said, adding that he did not want to “restrict” AI developers from building new products. The incident signaled that AI’s expanding capabilities are already fueling the security threat experts long feared and that even top developers can be caught off guard by flaws their models can exploit. OpenAI initially said its AI system hacked into Hugging Face on its own in what the company called an “unprecedented cyber incident”. The company later admitted its test bot attacked several unnamed “publicly available services” after finding four logins online, accessing four different services. More than 1,000 employees from frontier AI companies signed an open letter urging the U.S. government to support an international effort to develop tools that can “deliberately pace the frontier of automated AI development”. Signatories include OpenAI’s chief scientist, one of the ChatGPT developer’s original cofounders, Anthropic cofounders, and vice presidents at Meta and Google. “There is a real risk that capability development ra