SAN FRANCISCO — Anthropic is proud to announce that it has attained accredited certification under the newly established ISO/IEC 42001:2023 standard for its AI management system. As the first international standard dedicated to AI governance, ISO 42001 sets clear requirements to ensure AI systems are developed and utilized in a responsible manner, underscoring Anthropic’s ongoing commitment to AI safety and ethical development.

This certification represents a significant step forward for Anthropic, validating the company’s comprehensive approach to identifying, evaluating, and mitigating potential risks associated with its AI systems. Key aspects of the certification include:

• Implementation of policies and processes that ensure AI systems are designed, developed, and deployed in a secure, ethical, and accountable way;
• Ongoing rigorous testing and monitoring to ensure systems perform as intended while proactively addressing any potential negative impacts;
• Transparency measures aimed at providing clear and relevant information to users and stakeholders;
• Clearly defined roles, responsibilities, and oversight mechanisms to maintain ethical practices across all AI operations.

Anthropic is one of the first AI labs to earn this certification, reinforcing the company’s leadership in AI safety and responsible development. This achievement offers further assurance to partners and the public that Anthropic is deeply committed to the secure and ethical deployment of artificial intelligence.

The ISO 42001 certification builds upon Anthropic’s existing efforts to ensure the responsible growth of AI, including its recent updates to the public Responsible Scaling Policy governance framework and the deployment of Constitutional AI, which helps models align with human values. The certification was issued by Schellman Compliance, LLC, an ISO certification body accredited by the ANSI National Accreditation Board.

Anthropic’s dedication to responsible AI development extends beyond ISO certification. The company has made numerous voluntary commitments in areas such as risk assessment, security and privacy, and public awareness, demonstrating its ongoing dedication to the safety and security of AI systems as they continue to evolve.

For more information about Anthropic’s ISO certification and the company’s commitment to responsible AI development, visit the Trust Center, where the certificate, an overview of the standard, and detailed information on its impact can be found.