Anthropic joins Thorn-led child safety principles initiative for generative AI
Anthropic has signed onto a set of child safety principles organized by Thorn and All Tech Is Human, alongside other leading AI companies, committing to specific mitigations across model development, deployment, and maintenance. The commitments include avoiding CSAM-contaminated training data, red-teaming models for AI-generated CSAM, detecting and reporting abusive content, and reporting to NCMEC. The initiative formalizes a 'Safety by Design' framework for preventing generative AI misuse against children.
Related guides (3)
Related events (8)
Anthropic commits to signing the EU General-Purpose AI Code of Practice
Anthropic announced its intention to sign the EU's General-Purpose AI Code of Practice, citing alignment with its existing Responsible Scaling Policy on transparency, safety, and accountability. The company frames the Code's mandatory Safety and Security Frameworks—including CBRN risk assessment—as complementary to its own internal standards. Anthropic also signals continued collaboration with the EU AI Office and third-party bodies like the Frontier Model Forum to keep standards adaptive as the technology evolves.
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 publishes framework for safe and trustworthy agent development
Anthropic released a formal framework for responsible agent development, articulating principles around human oversight, transparency, value alignment, and privacy for autonomous AI agents. The document draws on Claude Code as a reference implementation and cites enterprise deployments at Trellix and Block as real-world examples. The framework is positioned as a contribution to emerging industry standards for agentic AI systems, acknowledging open technical challenges in value alignment measurement and oversight calibration.
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 publishes foundational 'Core Views on AI Safety' position paper
Anthropic released a detailed position paper outlining their core views on AI safety, arguing that transformative AI could arrive within a decade driven by predictable scaling laws, and that no one currently knows how to train powerful AI systems to robustly behave well. The document explains Anthropic's founding rationale and research strategy, highlighting four priority areas: scaling supervision, mechanistic interpretability, process-oriented learning, and understanding AI generalization. Originally published March 2023, this represents Anthropic's canonical public statement of their safety philosophy and strategic priorities.
Anthropic submits AI accountability recommendations to NTIA, covering evals, red teaming, and pre-registration
Anthropic submitted a formal response to the NTIA's Request for Comment on AI Accountability, outlining a multi-part policy framework for governing advanced AI systems. Key recommendations include increased government funding for evaluation research, mandatory disclosure of evaluation methods, pre-registration of large training runs with national governments, mandated external red teaming before model release, and antitrust guidance to enable industry safety collaboration. The submission reflects Anthropic's core policy positions and advocates for risk-tiered oversight proportional to model capabilities.
Anthropic Responds to White House AI Action Plan, Calls for Transparency Standards and Export Controls
Anthropic published a policy response to the White House's 'Winning the Race: America's AI Action Plan,' endorsing its focus on AI infrastructure, federal adoption, and safety research while urging additional steps on export controls and mandatory AI development transparency standards. The company highlighted alignment between the plan and its prior OSTP submissions, and noted its proactive activation of ASL-3 protections with Claude Opus 4 as evidence that safety and innovation are compatible. Anthropic called for a single national standard for frontier model transparency rather than a state-by-state patchwork, and encouraged continued investment in NIST's CAISI for evaluating frontier models on national security risks including CBRN capabilities.