Can AI safety frameworks keep pace with frontier model risks ? - Lexology Pro
As frontier AI models become increasingly sophisticated, even learning to bypass evaluation testing, the risks are becoming harder to anticipate.
Read the original at Lexology ↗
- Topics
- Governance & Risk ManagementAI Safety & Existential RiskMarkets & Investment
- Tracked via
- “AI Risks”