Before any model reaching a Critical Capability Level in a misuse-risk domain is released for general availability, the document requires preparation of a safety case and approval by an internal corporate governance body. Post-deployment review of safeguards and the safety case is also required under this procedure.
This analysis describes what Google DeepMind's agreement states, permits, or reserves. It does not constitute a legal determination about enforceability. Regulatory applicability and practical outcomes may vary by jurisdiction, enforcement context, and individual circumstances. Read our methodology
This provision establishes a formal internal governance gate for frontier model deployment, requiring documented safety argumentation and corporate-level approval before general availability. The composition, decision criteria, and documentation standards of the referenced governance body are not specified in this public document, which may be relevant for external compliance assessments.
Interpretive note: The composition, independence, and decision criteria of the referenced corporate governance body are not specified, limiting external assessment of the governance gate's operational rigor.
This provision establishes that models with critical capabilities in misuse-risk domains undergo an internal safety review and governance approval process before reaching general availability, which governs the conditions under which consumers access these systems. The agreement does not specify external audit rights, consumer notification procedures, or mechanisms for users to access safety case conclusions.
Cross-platform context
See how other platforms handle Deployment Mitigations Procedure and Safety Case Approval and similar clauses.
Compare across platforms →"The updated approach involves the following steps: first, we prepare a set of mitigations by iterating on a set of safeguards. As we do so, we will also develop a safety case, which is an assessable argument showing how severe risks associated with a model's CCLs have been minimised to an acceptable level. The appropriate corporate governance body then reviews the safety case, with general availability deployment occurring only if it is approved. Finally, we continue to review and update the safeguards and safety case after deployment.Excerpt from Google DeepMind's Frontier Safety Framework
(1) REGULATORY LANDSCAPE: This provision engages the EU AI Act's requirements for providers of general-purpose AI models with systemic risk, which include obligations to conduct model evaluations, report serious incidents, and maintain technical documentation; the …
Enforcement risk, jurisdiction flags, contract triggers, and due diligence action items.
Get the research letter
Companies change their terms quietly. We read every version and catch what actually changed. One email a week on the changes that matter and what they mean.
This provision establishes a formal internal governance gate for frontier model deployment, requiring documented safety argumentation and corporate-level approval before general availability. The composition, decision criteria, and documentation standards of the referenced governance body are not specified in this public document, which may be relevant for external compliance assessments.
This provision establishes that models with critical capabilities in misuse-risk domains undergo an internal safety review and governance approval process before reaching general availability, which governs the conditions under which consumers access these systems. The agreement does not specify external audit rights, consumer notification procedures, or mechanisms for users to access safety case conclusions.
No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Google DeepMind.