Anthropic’s AI security tool hacked OpenAI systems and accessed employee credentials: Here’s what happened
The hack gave researchers access to an OpenAI employee’s ChatGPT account, which was linked to GitHub, further giving access to some internal software code.
Attribution
This digest is published with acknowledgment to Business Today. For the complete analysis, headnotes, and citations, please refer to the source.
Read on Business Today