Responsible Scaling Policy
responsible-scaling-policy-d4e84707·20 events·first seen 15d agoAliases: Responsible Scaling Policy
Co-occurring entities
More like this (12)
Recent events (20)
Anthropic Releases Responsible Scaling Policy Version 3.0
Anthropic has published the third version of its Responsible Scaling Policy (RSP), a voluntary framework for mitigating catastrophic risks from increasingly capable AI systems. The update reflects two-plus years of experience with the original RSP, reinforcing what worked (ASL-3 safeguards activated in May 2025, industry adoption by OpenAI and Google DeepMind, informing early AI policy) while addressing shortcomings in accountability and transparency. The new version refines the AI Safety Level (ASL) framework and introduces new measures for decision-making transparency. Anthropic acknowledges that some elements of its original theory of change—particularly multilateral coordination and government action at higher capability thresholds—have not fully materialized as hoped.
Anthropic publishes Responsible Scaling Policy with AI Safety Level framework
Anthropic released its Responsible Scaling Policy (RSP), a formal framework of technical and organizational protocols for managing catastrophic risks from increasingly capable AI systems. The policy introduces AI Safety Levels (ASL-1 through ASL-5+), modeled on US biosafety level standards, requiring progressively stricter safety, security, and operational standards as models become more capable. Current Claude models are classified as ASL-2; ASL-3 triggers stricter deployment constraints including adversarial red-teaming requirements. The policy has been approved by Anthropic's board and is intended as a template for industry-wide adoption.
Anthropic publishes major update to Responsible Scaling Policy with new capability thresholds and ASL standards
Anthropic released a significant revision to its Responsible Scaling Policy (RSP), its risk governance framework for managing catastrophic risks from frontier AI. The update introduces two explicit capability thresholds—autonomous AI R&D and CBRN weapons uplift—that trigger mandatory upgrades to AI Safety Level (ASL) standards, with current models operating under ASL-2. New elements include safety-case-inspired documentation processes, internal governance stress-testing, and external expert input mechanisms, drawing on risk management practices from high-consequence industries like biosafety.
Anthropic reflects on Responsible Scaling Policy implementation and previews updated framework
Anthropic published a retrospective on operationalizing its Responsible Scaling Policy (RSP), originally released in summer 2023, sharing lessons learned and announcing an updated RSP is forthcoming. The post outlines five high-level commitments: establishing Red Line Capabilities, conducting Frontier Risk Evaluations, responding to Red Line Capabilities via an ASL-3 Standard, iteratively extending the policy toward ASL-4, and implementing Assurance Mechanisms. Key reflections include the difficulty of anticipating emergent capabilities in future models, expert disagreement on CBRN risk prioritization, and the value of quantitative threat modeling. Anthropic signals intent to move from voluntary commitments toward industry best practices and eventual regulation.
Dario Amodei's AI Safety Summit remarks detail Anthropic's Responsible Scaling Policy and ASL framework
Dario Amodei delivered prepared remarks at the UK AI Safety Summit (November 2023) explaining Anthropic's Responsible Scaling Policy (RSP), which was the first such policy published by a major AI lab. The RSP introduces AI Safety Levels (ASL-1 through ASL-4), modeled on biosafety level frameworks, with capability thresholds triggering mandatory safeguards before further training or deployment. Key implementation lessons include deep executive involvement, integrating RSP requirements into product roadmaps, and formal accountability through Anthropic's board and Long Term Benefit Trust. The remarks outline specific ASL-3 requirements around CBRN misuse prevention and security, and preview ASL-4 criteria involving near-human autonomy or becoming a primary source of global security threats.
Anthropic publishes policy brief calling for targeted AI regulation within 18 months
Anthropic published a policy position paper arguing that governments have an 18-month window to enact narrowly-targeted AI regulation before risks in cyber and CBRN domains become acute. The post cites rapid capability gains—SWE-bench scores rising from 1.96% to 49% in a year, GPQA scores approaching human expert level—as evidence that frontier models are approaching meaningful misuse thresholds. Anthropic also reviews its Responsible Scaling Policy as a model for adaptive, proportionate risk governance and calls for similar frameworks to be adopted industry-wide and codified in law.
Anthropic Publishes Frontier Compliance Framework for California's SB 53 Transparency in Frontier AI Act
Anthropic has released its Frontier Compliance Framework (FCF) in advance of California's SB 53 taking effect on January 1, 2026, which establishes the first mandatory frontier AI safety and transparency requirements in the US. The FCF covers risk assessment and mitigation for cyber, CBRN, and AI autonomy/control risks, tiered capability evaluation, model weight protection, and incident response. Anthropic frames the FCF as an evolution of its existing Responsible Scaling Policy, which will continue as a voluntary safety policy beyond regulatory minimums. The company also calls for a federal AI transparency framework with analogous requirements applied only to the largest frontier developers.
Anthropic activates ASL-3 safety protections for Claude Opus 4 launch
Anthropic has activated its AI Safety Level 3 (ASL-3) Deployment and Security Standards in conjunction with launching Claude Opus 4, marking the first time any Anthropic model has been deployed under ASL-3 rather than the baseline ASL-2. The activation is described as precautionary: Anthropic has not conclusively determined that Opus 4 crosses the ASL-3 capability threshold, but cannot rule it out due to continued improvements in CBRN-related knowledge. ASL-3 measures include Constitutional Classifiers to block end-to-end CBRN weapon development workflows and enhanced model-weight security against sophisticated non-state attackers. Claude Sonnet 4 was evaluated and cleared for ASL-2, and ASL-4 was ruled out for Opus 4.
Anthropic launches bug bounty program to stress-test ASL-3 Constitutional Classifiers
Anthropic launched an invite-only bug bounty program in partnership with HackerOne to find universal jailbreaks in its Constitutional Classifiers system before public deployment, offering up to $25,000 per verified vulnerability. The program targets CBRN-related safety bypasses on Claude 3.7 Sonnet and is part of Anthropic's work to meet its AI Safety Level-3 (ASL-3) Deployment Standard under its Responsible Scaling Policy. A follow-up update extended the program to test Constitutional Classifiers on the new Claude Opus 4 model and began accepting reports of universal jailbreaks found on public platforms. The initiative reflects Anthropic's structured approach to pre-deployment safety validation for increasingly capable models.
Anthropic launches initiative to fund third-party AI safety evaluations
Anthropic announced a funded initiative to source third-party evaluations measuring advanced AI capabilities and safety risks, with priority areas including cybersecurity, CBRN threats, model autonomy, national security risks, social manipulation, and misalignment. The initiative is tied to Anthropic's Responsible Scaling Policy and AI Safety Level (ASL) framework, aiming to address a gap between demand and supply of high-quality safety-relevant evals. Proposals are solicited via an application form, with Anthropic framing the effort as benefiting the broader AI safety ecosystem rather than just internal use.
Anthropic Endorses California SB 53 AI Safety Disclosure Bill
Anthropic has announced its endorsement of California Senate Bill 53, which would require large frontier AI developers to publish safety frameworks, release transparency reports before deploying powerful models, report critical safety incidents within 15 days, and provide whistleblower protections. The bill, authored by Senator Scott Wiener and informed by the Joint California Policy Working Group, takes a disclosure-based approach rather than prescriptive technical mandates, drawing lessons from the failed SB 1047. Anthropic frames the bill as formalizing practices already followed by major labs including Google DeepMind, OpenAI, and Microsoft, while creating a level playing field that prevents competitive pressure from eroding voluntary safety programs. Anthropic notes the bill's compute-based threshold (10^26 FLOPS) is an acceptable starting point but calls for future refinement as AI capabilities advance.
Anthropic Releases Computer Use Capability for Claude 3.5 Sonnet
Anthropic has launched a public beta of computer use for Claude 3.5 Sonnet, enabling the model to control a computer by interpreting screenshots and issuing pixel-level cursor and keyboard commands. The model achieves 14.9% on the OSWorld benchmark, roughly double the next-best AI model's 7.7%, though well below human-level performance of 70-75%. Anthropic trained the model on a small set of simple software tools and found it generalized rapidly to broader computer interaction. Safety analysis confirmed the capability remains at AI Safety Level 2, with prompt injection identified as a primary near-term risk.
Dario Amodei calls for stronger AI safety focus at Paris AI Action Summit
Anthropic CEO Dario Amodei issued a statement following the Paris AI Action Summit, expressing concern that the event underweighted critical issues including democratic leadership in AI, CBRN and autonomous-risk governance, and labor market disruption. Amodei forecasts that by 2026-2027 AI capabilities may be equivalent to 'a country of geniuses in a datacenter,' framing this as both an opportunity and an urgent governance challenge. He called for governments to enforce transparency of frontier lab safety plans, fund third-party evaluations, and monitor economic impacts—pointing to Anthropic's newly released Economic Index as a model. The statement also reaffirmed Anthropic's Responsible Scaling Policy as the first of its kind among frontier labs.
Anthropic advocates for third-party testing regime as core AI policy infrastructure
Anthropic published a policy position paper arguing that frontier AI systems require a third-party testing and oversight regime, distinct from self-governance approaches like their own Responsible Scaling Policy. The post outlines what such a regime should include: trusted third-party auditors, precisely scoped tests targeting only the most computationally intensive systems, and international coordination via shared standards and Mutual Recognition agreements. Anthropic acknowledges their RSP is insufficient alone because it relies on single private-sector actors, and calls for industry-wide mandatory testing that would eventually become a legal requirement for wide deployment.
Anthropic publishes structured harm assessment framework covering physical, psychological, economic, and societal impacts
Anthropic has released a policy document describing their evolving framework for assessing and mitigating AI harms across five dimensions: physical, psychological, economic, societal, and individual autonomy impacts. The framework complements their existing Responsible Scaling Policy and informs decisions on usage policies, red-teaming, detection, and enforcement. Concrete examples include safeguards for computer use capabilities (fraud, phishing) and a reported 45% reduction in unnecessary refusals in Claude 3.7 Sonnet through improved handling of ambiguous prompts. Anthropic frames this as a work-in-progress and invites collaboration from the broader AI ecosystem.
Anthropic responds to California Governor Newsom's AI working group draft report
Anthropic published a formal response to the California Governor's Working Group on AI Frontier Models draft report, endorsing its emphasis on transparency and evidence-based policy. Anthropic argues that light-touch mandatory disclosure of safety and security practices would be beneficial without impeding innovation, noting that current voluntary practices are uneven across frontier labs. The response also references Anthropic's Responsible Scaling Policy and Economic Index as examples of existing transparency efforts, and signals urgency given Anthropic's view that powerful AI systems may arrive as early as end of 2026.
Anthropic proposes federal AI transparency framework with mandatory Secure Development Frameworks and system cards
Anthropic published a policy proposal calling for a targeted AI transparency framework applicable at federal, state, or international levels, targeting only the largest frontier AI developers (suggested thresholds: ~$100M annual revenue or ~$1B R&D/capex). The framework would require covered developers to publicly disclose a Secure Development Framework covering CBRN and misalignment risks, publish system cards at deployment, self-certify compliance, and face legal liability for false statements. The proposal is explicitly lightweight and flexible, designed to avoid prescriptive standards while creating accountability mechanisms and whistleblower protections during the period before comprehensive safety standards are established.
Anthropic makes Claude 3 Haiku and Sonnet available to US Intelligence Community and AWS GovCloud
Anthropic has made Claude 3 Haiku and Claude 3 Sonnet available via AWS Marketplace for the US Intelligence Community and AWS GovCloud, marking a significant expansion into government deployment. The company has crafted contractual exceptions to its general Usage Policy to permit legally authorized foreign intelligence analysis, including combating human trafficking and identifying covert influence campaigns, while maintaining restrictions on disinformation, weapons design, and malicious cyber operations. The deployment is currently limited to ASL-2 models under Anthropic's Responsible Scaling Policy. Anthropic also notes prior pre-release access to Claude 3.5 Sonnet was provided to the UK AI Safety Institute for pre-deployment testing.
Anthropic introduces computer use capability, upgraded Claude 3.5 Sonnet, and Claude 3.5 Haiku
Anthropic announced three major developments: an upgraded Claude 3.5 Sonnet with significant coding improvements (SWE-bench Verified rising from 33.4% to 49.0%, surpassing all publicly available models including reasoning models), a new Claude 3.5 Haiku that matches Claude 3 Opus performance at Haiku-tier speed, and a public beta of 'computer use' — a capability allowing Claude to control computers by viewing screens, moving cursors, clicking, and typing. Computer use is available via the Anthropic API, Amazon Bedrock, and Google Cloud Vertex AI, with early adopters including Replit, The Browser Company, and Cognition. Both safety institutes (US AISI and UK AISI) conducted pre-deployment testing, and the model was assessed as remaining within ASL-2 under Anthropic's Responsible Scaling Policy.
Anthropic achieves ISO/IEC 42001:2023 certification for AI management systems
Anthropic has received accredited certification under ISO/IEC 42001:2023, the first international standard for AI governance and management systems, issued by Schellman Compliance LLC. The certification covers Anthropic's policies, testing, monitoring, transparency measures, and oversight structures for responsible AI development. Anthropic claims to be among the first frontier AI labs to achieve this certification, positioning it as external validation of their safety commitments alongside existing frameworks like their Responsible Scaling Policy and Constitutional AI.