Print-friendly view · use your browser's Save as PDF option (Cmd/Ctrl-P) to attach this article to a brief.

Meta Frontier AI Framework

META-FRONTIER-2024 · US · voluntary code

Source: https://policywindow.org/wiki/meta-frontier-ai-framework

Generated 2026-05-30T22:11:50 UTC

Summary

Meta's open-weight-frontier governance posture. Categorises frontier models into 'high risk' + 'critical risk' tiers; the framework's distinctive feature is its explicit defence of open-weight release as a governance posture (vs. the closed-model stance of Anthropic / OpenAI / DeepMind). Pre-release threat modelling + post-release monitoring; commits to halt training if critical-risk threshold reached without mitigations. Seoul Frontier AI Safety Commitments signatory.

At a glance

Adopted: 2024-02-02
Effective: 2024-02-02
Status: in force
Primary source: Meta Frontier AI Framework (Feb 2024)
Source URL: https://ai.meta.com/responsible-ai/

How to cite this article

APA

Policy Window. (2024). Meta Frontier AI Framework [Wiki article — Instrument]. https://policywindow.org/wiki/meta-frontier-ai-framework

Chicago

Policy Window. 2024. "Meta Frontier AI Framework." Wiki article (Instrument). https://policywindow.org/wiki/meta-frontier-ai-framework.

Harvard

Policy Window (2024) 'Meta Frontier AI Framework', Wiki article — Instrument, available at: https://policywindow.org/wiki/meta-frontier-ai-framework.

OSCOLA

Policy Window, 'Meta Frontier AI Framework' (Wiki article — Instrument, 2024) <https://policywindow.org/wiki/meta-frontier-ai-framework> accessed [date].

BibTeX

@misc{policywindow-meta-frontier-ai-framework,
  title  = {Meta Frontier AI Framework},
  author = {Policy Window},
  year   = {2024},
  howpublished = {Meta Frontier AI Framework (Feb 2024)},
  url    = {https://policywindow.org/wiki/meta-frontier-ai-framework},
  note   = {Primary source: https://ai.meta.com/responsible-ai/}
}