Google DeepMind Frontier Safety Framework Version 1.0

• Authority: Private-sector companies
• Collections: Corporate policies and commitments;Editors' Picks
• Most recent activity: Defunct
• Most recent activity date: 2025-02-04

Risk subdomain coverage

Good coverage — Comprehensive governance measures or mitigations with clear procedures relating to the risk subdomain are explicitly described.

• AI system security vulnerabilities
• Cyberattacks, weapon development or use
• Competitive dynamics
• Governance failure
• AI pursuing its own goals
• AI possessing dangerous capabilities
• Lack of capability or robustness

Minimal coverage — Brief mention but no specific mitigations, recommendations or governance measures relating to the risk subdomain are described.

• Disinformation, surveillance, and influence at scale
• Fraud, scams, and targeted manipulation
• Multi-agent risks

All other risk subdomains had no coverage in this document.

Risk Coverage Summary

The document has good coverage of approximately 10-12 subdomains, with strong focus on malicious actors (4.1, 4.2, 4.3), AI system security (2.2), competitive dynamics (6.4), AI safety failures (7.1, 7.2, 7.3), and governance structures (6.5). Coverage is concentrated in security, misuse prevention, and AI safety domains, with minimal attention to discrimination, privacy, misinformation, or socioeconomic impacts.

Sector coverage

Good coverage — Comprehensive governance measures or mitigations with clear procedures relating to the sector are explicitly described.

• Information
• Scientific Research & Development Services

All other sectors had no coverage in this document.

Sectors Summary

This is an internal corporate policy document governing Google DeepMind's own AI development activities. As an AI research and development company, Google DeepMind operates primarily in the Information sector (AI/ML development) and Scientific Research and Development Services sector. The document does not regulate external sectors but rather governs the company's internal processes for developing frontier AI models.

Legislative Status

Other

Legislative Status Summary

This is an internal corporate policy document establishing Google DeepMind's own protocols for managing frontier AI risks. It uses voluntary language ('we will', 'we aim to') and lacks external enforcement mechanisms, but goes beyond typical soft law by establishing specific internal commitments and procedures.

Actors

Proposer

• Entity Types: AI_Developer
• Entities: Google DeepMind
• Summary: The document is authored by Google DeepMind and describes 'our first version of a set of protocols' and 'our existing suite of AI responsibility and safety practices', clearly indicating Google DeepMind as the proposer of this framework.

Target

• Entity Types: AI_Developer
• Entities: Google DeepMind
• Summary: The framework applies to Google DeepMind's own frontier model development and deployment activities. The document consistently uses 'our models', 'our evaluators', and 'our risk assessment' indicating this is an internal policy governing the company's own AI development.

Enforcer

• Entity Types: AI_Developer
• Entities: Google DeepMind
• Summary: Google DeepMind enforces this framework on itself through internal processes. The document describes self-enforcement mechanisms including evaluation protocols, mitigation requirements, and the authority to halt deployment or development.

Monitor

• Entity Types: AI_Developer, AI_Governance_Actor
• Entities: Google DeepMind; external parties (unspecified); independent third parties (unspecified); relevant stakeholder bodies (unspecified)
• Summary: Primary monitoring is conducted by Google DeepMind itself through periodic evaluations and red-teaming. The document also mentions future involvement of external authorities and independent third parties in monitoring and oversight, though specific entities are not named.

AI Lifecycle Stages

Lifecycle stages covered by the documents, as defined in the OECD Framework for the Classification of AI Systems

• Plan and Design
• Build and Use Model
• Verify and Validate
• Deploy
• Operate and Monitor

AI Lifecycle Summary

The document comprehensively covers Build and Use Model, Verify and Validate, Deploy, and Operate and Monitor stages. It focuses extensively on model evaluation, testing protocols, deployment mitigations, and ongoing monitoring of frontier models. There is minimal coverage of Plan and Design, and no explicit coverage of data collection and processing stages.

Technical Scope

Terms relating to technical scope that are mentioned or implicitly referenced in the document

• AI Models
• Foundation Models
• Frontier AI
• AI Systems
• Compute Threshold
• Open Weight / Open Source

Technical Scope Summary

The document explicitly focuses on 'frontier models' and 'foundation models' as its primary scope. It does not use the terms general purpose AI, task-specific AI, or predictive AI. It mentions compute thresholds indirectly through '6x in effective compute' evaluation intervals. The framework addresses both model weights and deployed systems, with specific attention to preventing open release of model weights.