AI arms race in line for a reckoning after OpenAI hacking incident
AI disclosure
Summary
Aggressive training techniques sharpens threat of bad behavior by leading models.