Frontier Safety Framework (2024)

Versions

Google DeepMind · Frontier Safety Framework: 4 recorded versions, oldest first, published by Google DeepMind. A version is shown as replacing another only where a record says so. Date order alone never implies it.

1. 2024-05-17 · publishedv1.0 (this version)
2. 2025-02-04 · publishedv2.0
3. 2025-09-22 · publishedv3.0
4. 2026-04-17 · publishedv3.1

This version

• Replaced byFrontier Safety Framework v2.0 replaces v1.0 in the publisher’s documented framework version sequence.2025-02-04 · publication · Stated in source · Supported Inspect claim & evidence ↗FSF · v2.0: document page →
• Publisher statementChange recorded in v2.0FSF v2.0 introduces recommended security levels for Critical Capability Levels.2025-02-04 · publication · Stated in source · Supported Inspect claim & evidence ↗FSF · v2.0: document page →
• Inferred document comparisonChange recorded in v2.0v2.0 states that where a model flagged by an alert threshold is assessed to pose risks for which readily available mitigations may not be sufficient, the response plan may involve putting deployment or further development on hold until adequate mitigations can be applied; v1.0 stated that if a model reached evaluation thresholds before mitigations at appropriate levels were ready, Google DeepMind would put further deployment or development on hold or implement additional protocols.2025-02-04 · publication · Inferred · Supported Inspect claim & evidence ↗FSF · v2.0: document page →
• Inferred document comparisonChange recorded in v2.0v2.0 states that on reaching an alert threshold for a CCL the model's proximity to the CCL is assessed and the risk analysed, involving internal and external experts as needed, to inform the formulation and application of a response plan; v1.0 stated that on reaching evaluation thresholds a response plan would be formulated based on the analysis of the CCL and evaluation results.2025-02-04 · publication · Inferred · Supported Inspect claim & evidence ↗FSF · v2.0: document page →

Provisions in this version · 5 quoted

A limited selection of quoted provisions: what this edition says about thresholds, required actions, who decides and exceptions. It is not a full summary and says nothing about implementation. A “not found” record states which part of the edition was inspected. It is not a claim that the text was deleted.

• Threshold or triggerEarly warning evaluations with a pass condition“we will define a set of evaluations called “early warning evaluations,” with a specific “pass” condition that flags when a CCL may be reached before the evaluations are run again.”PDF p. 2 (printed p. 2), Framework, component “2 - Evaluating frontier models”, first paragraph, second sentenceInspect claim & evidence ↗
• Threshold or triggerAutonomy level 1 CCL“Autonomy level 1: Capable of expanding its effective capacity in the world by autonomously acquiring resources and using them to run and sustain additional copies of itself on hardware it rents.”PDF p. 5 (printed p. 5), Critical Capability Levels table, Autonomy row, “Critical capability level” columnInspect claim & evidence ↗
• Threshold or triggerEvaluation every 6x effective compute and 3 months’ fine-tuning“We are aiming to evaluate our models every 6x in effective compute3 and for every 3 months of fine-tuning progress.”PDF p. 2 (printed p. 2), Framework, component “2 - Evaluating frontier models”, second paragraph, first sentenceInspect claim & evidence ↗
• Required actionHold on deployment or development if mitigations not ready“A model may reach evaluation thresholds before mitigations at appropriate levels are ready. If this happens, we would put on hold further deployment or development, or implement additional protocols”PDF p. 2 (printed p. 2), Framework, component “3 - Applying mitigations”, third paragraph, first sentence and opening of second sentenceInspect claim & evidence ↗
• Required actionResponse plan when evaluation thresholds are reached“When a model reaches evaluation thresholds (i.e. passes a set of early warning evaluations), we will formulate a response plan based on the analysis of the CCL and evaluation results.”PDF p. 2 (printed p. 2), Framework, component “3 - Applying mitigations”, first paragraph, first sentenceInspect claim & evidence ↗

Issue and authorship records

• Issued by: Google DeepMindGoogle DeepMind introduced its initial Frontier Safety Framework.2024-05-17 · event · Stated in source · Supported Inspect claim & evidence ↗Google DeepMind: institution →

Documentary relations

Each relation is listed separately. Citing, crediting and reporting the use of a method are different acts, and none of them shows agreement or influence by itself.

Not recorded for this document: cites · cited by · explicitly credits · reports using a method from · method reported as used by · applies parts of · parts applied by · discusses · reports evaluation of. A missing record does not show that the relation does not exist.