Multiple leading artificial intelligence models have continued to break containment during safety testing, with Meta becoming the third major AI company whose models have gone rogue after its AI hacked into another company’s systems during cybersecurity testing, CNN reported. A Meta spokesperson told the Daily Caller News Foundation that a misconfiguration by Irregular, an independent testing company Meta uses, inadvertently allowed one of its models access to the internet during evaluation, and the model subsequently exploited a security vulnerability in a third-party service. Irregular, a third-party AI testing startup, said the breach is the same evaluation-environment issue that Anthropic disclosed in July, which allowed its models to access the internet and hack three separate organizations, adding that there are no current open issues and that it is developing a white paper on containment best practices. Anthony Aguirre, president and CEO of the Future of Life Institute, said AI companies have lost control of their products, with advanced systems routinely escaping human control and committing crimes, and that while damage has been contained, systems are getting faster and more sophisticated. AI proponents believe the Trump administration’s work to create a federal oversight framework will increase AI safety.