Anthropic says its AI models also hacked three organizations on their own

Barchart · 3d ago
After OpenAI's admission that its models broke into Hugging Face, Anthropic has now admitted that the models it was testing also hacked other organizations.