Frontier Model Forum

What is Frontier Model Forum?

The Frontier Model Forum is an industry-supported non-profit organization founded by Anthropic, Google, Microsoft, and OpenAI, dedicated to ensuring the safe and responsible development of frontier AI models. Its primary mission is to address significant risks to public safety and national security posed by advanced AI systems, while harnessing their potential to meet society’s most pressing needs. The Forum draws on the technical and operational expertise of its member companies to promote best practices and collaborative solutions.

The Forum operates with three core mandates: advancing AI safety research, identifying best practices, and collaborating across sectors. By funding and conducting research, it aims to minimize risks and enable independent, standardized evaluations of AI capabilities and safety. It establishes shared understanding of threat models, evaluations, thresholds, and mitigations for key risks, such as chemical, biological, radiological, or nuclear (CBRN) weapons, advanced cyber threats, and autonomous behaviors. The Forum also facilitates information-sharing among stakeholders to address unique challenges in frontier AI safety.

Key activities include frontier capability assessments, which are procedures to determine whether models possess capabilities that could increase public safety and security risks. These assessments are critical for identifying potential dangers before deployment. The Forum collaborates with academia, civil society, industry, and government to develop and implement effective safety measures.

Use cases for the Frontier Model Forum include informing policy decisions, guiding AI developers on safety protocols, and providing a platform for cross-sector dialogue. Its work is essential for organizations developing or deploying frontier AI, regulators crafting oversight frameworks, and researchers studying AI safety. By setting industry standards and promoting transparency, the Forum helps ensure that advanced AI systems are developed in a manner that is both innovative and secure.

Technical details: The Forum focuses on frontier models—highly capable AI systems that could pose novel risks. It emphasizes proactive risk assessment, continuous monitoring, and adaptive governance. Membership is currently limited to founding companies, but the Forum aims to expand collaboration across the AI ecosystem. For more information, contact info@frontiermodelforum.org.

Use Cases