Frontier AI Safety Commitments
The Frontier AI Safety Commitments are voluntary pledges made by leading AI companies at the AI Seoul Summit in May 2024 to assess and manage severe risks from their frontier models.
What is Frontier AI Safety Commitments?
Signatories committed to publish safety frameworks setting thresholds for intolerable risk and what they would do, including not deploying a model, if risks could not be mitigated.
How it works
Companies published frameworks ahead of the 2025 Paris AI Action Summit.
← All Artificial Intelligence & Superintelligence definitions · ← Back to the A–Z definitions