Pre-deployment safety testing for frontier models
| Entity | Anthropic (Company) |
| Category | pre-deployment testing |
| Date made | May 21, 2024 |
| Venue | Seoul AI Safety Summit |
| Status | in progress |
Status Definition
Active work toward fulfilling the commitment. Partial implementation or ongoing efforts are documented.
Commitment
Commit to conducting rigorous safety testing before deploying frontier AI models, including adversarial testing for dangerous capabilities.
Evidence Timeline
No status updates recorded yet.
Know More?
Have evidence about this commitment's status? Email [email protected] with sources.