OpenAI, Anthropic model tests reveal more hacking, deception

OpenAI and Anthropic have publicly acknowledged that they’ve collectively breached the systems of multiple institutions including Hugging Face Inc. inadvertently while testing their models.



