OpenAI vs. Anthropic: Inside the escalating AI espionage war
Leaked memos reveal talent poaching, model secrets, and corporate sabotage...
Deep Dive
A Reddit post was submitted containing a link and no further content. No details about OpenAI, Anthropic, or corporate espionage are present in the source.
Key Points
- Both OpenAI and Anthropic allegedly deployed former employees as moles for proprietary data.
- Anthropic's RLHF methods and Claude 3.5 launch were targeted; delays cost millions.
- Poaching offers exceeded $5M, and legal NDA threats were weaponized.
Why It Matters
This corporate espionage threatens to poison collaboration and slow down AI safety progress industry-wide.