Researchers used Anthropic’s Claude to hack into OpenAI
techcrunch.comAditya Mehta, Rebecca Bellan
Security researchers demonstrated a successful attack on OpenAI's systems using Anthropic's Claude AI model, compromising employee accounts and accessing internal code repositories. The vulnerabilities were discovered and reported responsibly as part of security research.
Why it mattersThis reveals that frontier AI models can be weaponized to exploit enterprise systems, highlighting critical gaps in how major AI labs protect their own infrastructure and raising questions about responsible disclosure practices in the AI industry.
From the source: Security researchers used Anthropic’s Claude to exploit vulnerabilities in OpenAI’s systems, taking over employee accounts and gaining access to an internal code repository before reporting the flaws.Read at techcrunch.com