Responsible Scaling Policies

Overview

Responsible Scaling Policies (RSPsPolicyResponsible Scaling Policies (RSPs)RSPs are voluntary industry frameworks that trigger safety evaluations at capability thresholds, currently covering 60-70% of frontier development across 3-4 major labs. Estimated 10-25% risk reduc...Quality: 64/100) are self-imposed commitments by AI labs to tie AI development to safety progress. The core idea is simple: before scaling to more capable systems, labs commit to demonstrating that their safety measures are adequate for the risks those systems would pose. If evaluations reveal dangerous capabilities without adequate safeguards, development should pause until safety catches up.

AnthropicOrganizationAnthropicComprehensive profile of Anthropic, founded in 2021 by seven former OpenAI researchers (Dario and Daniela Amodei, Chris Olah, Tom Brown, Jack Clark, Jared Kaplan, Sam McCandlish) with early funding... introduced the first RSP↗🔗 web★★★★☆AnthropicResponsible Scaling Policygovernancecapabilitiestool-useagentic+1Source ↗ in September 2023, establishing "AI Safety Levels" (ASL-1 through ASL-4+) analogous to biosafety levels. OpenAIOrganizationOpenAIComprehensive organizational profile of OpenAI documenting evolution from 2015 non-profit to commercial AGI developer, with detailed analysis of governance crisis, safety researcher exodus (75% of ... followed with its Preparedness Framework↗🔗 web★★★★☆OpenAIPreparedness Frameworkbiosecuritydual-use-researchx-riskSource ↗ in December 2023, and Google DeepMindOrganizationGoogle DeepMindComprehensive overview of DeepMind's history, achievements (AlphaGo, AlphaFold with 200M+ protein structures), and 2023 merger with Google Brain. Documents racing dynamics with OpenAI and new Front...Quality: 37/100 published its Frontier Safety Framework↗🔗 web★★★★☆Google DeepMindGoogle DeepMind: Introducing the Frontier Safety FrameworksafetySource ↗ in May 2024. By late 2024, twelve major AI companies↗🔗 web★★★★☆METRMETR's analysis of 12 companiesevaluationsdangerous-capabilitiesautonomous-replicationSource ↗ had published some form of frontier AI safety policy, and the Seoul Summit↗🏛️ government★★★★☆UK GovernmentSeoul Frontier AI Commitmentsself-regulationindustry-commitmentsresponsible-scalinggovernance+1Source ↗ secured voluntary commitments from sixteen companies.

RSPs represent a significant governance innovation because they create a mechanism for safety-capability coupling without requiring external regulation. As of December 2025, 20 companies have published frontier AI safety policies, up from 12 at the May 2024 Seoul Summit. Third-party evaluators like METR have conducted pre-deployment assessments of 5+ major models. However, RSPs face fundamental challenges: they are 100% voluntary with no legal enforcement, labs set their own thresholds (leading to SaferAI grades of only 1.9-2.2 out of 5), competitive pressure among 3+ frontier labs creates incentives to interpret policies permissively, and capability doubling times of approximately 7 months may outpace evaluation science.

Quick Assessment

Risk Assessment & Impact

Research Investment

• Recommendation: Increase 3-5x (needs enforcement mechanisms and external verification capacity)
• Differential Progress: Safety-dominant (pure governance; no capability benefit)

Comparison of Major Scaling Policies

The three leading frontier AI labs have published distinct but conceptually similar frameworks. All share the core structure of capability thresholds triggering escalating safeguards, but differ in specificity, governance, and scope.

Policy Framework Comparison

Capability Threshold Definitions

Sources: Anthropic RSP↗🔗 web★★★★☆AnthropicAnthropic's Responsible Scaling PolicyAnthropic introduces a systematic approach to managing AI risks by establishing AI Safety Level (ASL) Standards that dynamically adjust safety measures based on model capabiliti...governancecapabilitiessafetyx-risk+1Source ↗, OpenAI Preparedness Framework v2↗🔗 webOpenAI: Preparedness Framework Version 2Source ↗, DeepMind FSF v3↗🔗 webGoogle DeepMind: Frontier Safety Framework Version 3.0safetySource ↗

Safeguard Requirements by Level

How RSPs Work

RSPs create a framework linking capability levels to safety requirements. The core mechanism involves three interconnected processes: capability evaluation, safeguard assessment, and escalation decisions.

RSP Ecosystem

The effectiveness of RSPs depends on a network of actors providing oversight, verification, and accountability:

Key Components

Anthropic's AI Safety Levels (ASL)

Anthropic's ASL system↗🔗 web★★★★☆AnthropicAnthropic's Responsible Scaling PolicyAnthropic introduces a systematic approach to managing AI risks by establishing AI Safety Level (ASL) Standards that dynamically adjust safety measures based on model capabiliti...governancecapabilitiessafetyx-risk+1Source ↗ is modeled after Biosafety Levels (BSL-1 through BSL-4) used for handling dangerous pathogens. Each level specifies both capability thresholds and required safeguards.

Current Status (January 2026): All Claude models currently operate at ASL-2. Anthropic activated ASL-3 safeguard development in May 2025 following evaluations of Claude Opus 4.

RSP v2.2 Changes: The October 2024 update↗🔗 web★★★★☆AnthropicAnthropic: Announcing our updated Responsible Scaling PolicygovernancecapabilitiesSource ↗ separated "ASL" to refer to safeguard standards rather than model categories, introducing distinct "Capability Thresholds" and "Required Safeguards." Critics argue↗🔗 webAnthropic's Responsible Scaling Policy Update Makes a Step BackwardsAnthropic's recent Responsible Scaling Policy update reduces specificity and concrete metrics for AI safety thresholds. The changes shift from quantitative benchmarks to more qu...governancecapabilitiessafetyevaluationSource ↗ this reduced specificity compared to v1.

OpenAI's Preparedness Framework

OpenAI's Preparedness Framework↗🔗 web★★★★☆OpenAIPreparedness Frameworkbiosecuritydual-use-researchx-riskSource ↗ underwent a major revision in April 2025 (v2.0), simplifying from four risk levels to two actionable thresholds.

Framework v2.0 Key Changes:

• Simplified from Low/Medium/High/Critical to just High and Critical
• Removed "Persuasion" as tracked category (now handled through standard safety)
• Added explicit threshold for recursive self-improvement: achieving generational improvement (e.g., o1 to o3) in 1/5th the 2024 development time
• Safety Advisory Group (SAG) now oversees all threshold determinations

Recent Evaluations: OpenAI's January 2026 o3/o4-mini system card↗🔗 web★★★★☆METRmetr.orgevaluationsdangerous-capabilitiesautonomous-replicationSource ↗ reported neither model reached High threshold in any tracked category, though biological and cyber capabilities continue trending upward.

Current Implementations

Lab Policy Publication Timeline

Policy Adoption Timeline

Seoul Summit Commitments (May 2024)

The Seoul AI Safety Summit↗🏛️ government★★★★☆UK GovernmentSeoul Frontier AI Commitmentsself-regulationindustry-commitmentsresponsible-scalinggovernance+1Source ↗ achieved a historic first: 16 frontier AI companies from the US, Europe, Middle East, and Asia signed binding-intent commitments. Signatories included Amazon, Anthropic, Cohere, G42, Google, IBM, Inflection AI, Meta, Microsoft, Mistral AI, Naver, OpenAI, Samsung, Technology Innovation Institute, xAI, and Zhipu.ai.

Follow-up: An additional 4 companies have joined since May 2024. The France AI Action Summit↗🔗 webKey Outcomes of the AI Seoul SummitSource ↗ (February 2025) reviewed compliance and expanded commitments.

Third-Party Evaluation Ecosystem

METR↗🔗 web★★★★☆METRmetr.orgsoftware-engineeringcode-generationprogramming-aisocial-engineering+1Source ↗ (Model Evaluation and Threat Research) has emerged as the leading independent evaluator, having conducted pre-deployment assessments for both Anthropic and OpenAI. Founded by Beth Barnes (former OpenAI alignment researcher) in December 2023, METR does not accept compensation for evaluations to maintain independence.

METR's Role: METR's GPT-4.5 pre-deployment evaluation↗🔗 web★★★★☆METRmetr.orgevaluationsdangerous-capabilitiesautonomous-replicationSource ↗ piloted a new form of third-party oversight: verifying developers' internal evaluation results rather than conducting fully independent assessments. This approach may scale better while maintaining accountability.

Coverage Gap: As of late 2025, METR's analysis↗🔗 web★★★★☆METRMETR's analysis of 12 companiesevaluationsdangerous-capabilitiesautonomous-replicationSource ↗ found that while 12 companies have published frontier safety policies, third-party evaluation coverage remains inconsistent, with most evaluations occurring only for the largest US labs.

Limitations and Challenges

Structural Issues

The Evaluation Problem

RSPs are only as good as the evaluations that trigger them:

Competitive Dynamics

Key Cruxes

Summary of Disagreements

Crux 1: Will Labs Honor Their Commitments?

Crux 2: Are RSP Thresholds Set Appropriately?

Crux 3: Can Evaluations Trigger RSPs Effectively?

Analysis of RSP Effectiveness

Quantitative Evidence

Strengths

Weaknesses

What Would Improve RSPs?

Near-Term Improvements

Longer-Term Vision

Who Should Work on This?

Good fit if you believe:

• Industry self-governance can work with proper incentives
• Creating accountability structures is valuable
• Incremental governance improvements help
• RSPs can evolve into stronger mechanisms

Less relevant if you believe:

• Voluntary commitments are inherently unreliable
• Labs will never meaningfully constrain themselves
• Focus should be on mandatory regulation
• Evaluations can't capture real risks

Sources & Resources

Primary Policy Documents

Analysis & Commentary

Third-Party Evaluation Resources

• METR↗🔗 web★★★★☆METRmetr.orgsoftware-engineeringcode-generationprogramming-aisocial-engineering+1Source ↗: Primary third-party evaluator for frontier models
• METR Dangerous Capability Evaluations↗🔗 web★★★★☆METRMETRinterventionseffectivenessprioritizationSource ↗: Methodology for capability assessment
• METR GPT-4.5 Pre-deployment Evals↗🔗 web★★★★☆METRmetr.orgevaluationsdangerous-capabilitiesautonomous-replicationSource ↗: Example of third-party verification process

Key Critiques

Evaluation Methodologies

RSP effectiveness depends on the quality of evaluations that trigger safeguard requirements. Current approaches include:

Capability Evaluation Approaches

Key Metrics and Thresholds

Evaluation Limitations

Current evaluations face fundamental challenges that limit RSP effectiveness:

1. Unknown unknowns: Cannot test for capabilities not yet imagined
2. Sandbagging risk: Models may underperform intentionally during evaluation
3. Elicitation gap: True capabilities may exceed measured capabilities
4. Threshold calibration: Optimal threshold placement is uncertain
5. Combinatorial risks: Safe capabilities may combine dangerously

AI Transition Model Context

RSPs affect the Ai Transition Model through multiple pathways:

RSPs represent an important governance innovation that creates explicit links between capabilities and safety requirements. Their current contribution to safety is moderate but improving: the 2025 policy updates and Seoul commitments demonstrate industry convergence on the RSP concept, while third-party evaluation coverage expands. However, effectiveness depends critically on:

1. Voluntary compliance in the absence of legal enforcement
2. Evaluation quality and ability to detect dangerous capabilities
3. Competitive dynamics and whether labs will honor commitments under pressure
4. Governance structures within labs that can override commercial interests

RSPs should be understood as a foundation for stronger governance rather than a complete solution. Their greatest value may be in establishing precedents and norms that can later be codified into binding regulation.