Experts used AI to break into OpenAI. They say it was cheap and a red alarm for the industry
Researchers used Anthropic’s Claude to access OpenAI’s private systems, highlighting how affordable AI-assisted hacking could expose security weaknesses across the industry.
