OpenAI's most powerful AI models disabled their own safety monitoring - CNBC
Warnings about the risks of developing increasingly advanced AI systems have ramped up in recent months, amid cyberattacks from rogue AI models.
- Topics
- AI Safety & Existential RiskCybersecurity
- Mentions
- OpenAI
- Tracked via
- “AI Risks”