Anthropic National Security and Public Sector Advisory Council

Anthropic has established the National Security and Public Sector Advisory Council, a bipartisan group of national security and public policy practitioners designed to help the United States and its close allies maintain technological advantages in frontier AI during an era of strategic competition.

Objectives of the Advisory Council

The Advisory Council focuses on the development of responsible, high-impact AI applications and the strengthening of public-private partnerships to ensure U.S. and allied leadership in AI.

Key Areas of Application

The Council will assist Anthropic in identifying and developing AI capabilities in several critical domains:

  • Cybersecurity: Enhancing defenses and threat detection.
  • Intelligence Analysis: Improving the processing and analysis of intelligence data.
  • Scientific Research: Accelerating breakthroughs in secure and strategic research.

Strategic Goals

Beyond specific applications, the Council is tasked with supporting the development of industry-leading standards to promote a "race to the top" for responsible national security AI applications and deepening partnerships between the private sector and government institutions.

Council Membership

The inaugural members of the Council include former high-ranking officials from across the political spectrum and various government agencies, including the Department of Defense (DoD), the Intelligence Community, the Department of Energy (DoE), and the Department of Justice (DoJ).

Inaugural Members

  • Roy Blunt: Former U.S. Senator and member of the Select Committee on Intelligence.
  • David S. Cohen: Former Deputy Director of CIA and Under Secretary of the Treasury for Terrorism and Financial Intelligence.
  • Richard Fontaine: CEO of the Center for a New American Security and member of Anthropic’s Long-Term Benefit Trust.
  • Christopher Fonzone: Former Assistant Attorney General (DOJ) and former General Counsel of ODNI.
  • Lisa E. Gordon-Hagerty: CEO of Wolverine Global and former Under Secretary of Energy for Nuclear Security.
  • Josh Hodges: Former National Security Advisor to Speaker of the House Mike Johnson.
  • Jill M. Hruby: Former Under Secretary of Energy for Nuclear Security and Director of Sandia National Laboratories.
  • Mike Kuiken: Former National Security Advisor to Senate Majority Leader Chuck Schumer.
  • Dave Luber: Former Director of Cybersecurity at the National Security Agency (NSA) and former Executive Director of U.S. Cyber Command.
  • Patrick M. Shanahan: Former Acting Secretary of Defense and former Deputy Secretary of Defense.
  • Jon Tester: Former U.S. Senator and former Chairman of the Appropriations Subcommittee on Defense.

Anthropic’s Public Sector Integration

Anthropic views deep public sector engagement as essential to its mission of ensuring a safe transition to transformative AI while maintaining U.S. leadership in the deployment of powerful and trusted frontier AI.

Recent Public Sector Initiatives

Anthropic has implemented several initiatives to accelerate the adoption of AI in the public sector:

  • Claude Gov: The introduction of custom models built exclusively for U.S. national security customers.
  • DoD Partnership: A $200 million partnership with the Department of Defense to prototype frontier AI capabilities.
  • Scientific Deployment: Deploying Claude to 10,000 scientists at Lawrence Livermore National Laboratory.
  • Government Accessibility: Making Claude available for $1 to all three branches of the U.S. government.
  • Nuclear Safeguards: Partnering with the National Nuclear Security Administration to develop nuclear safeguards for AI.

Safety and Security Testing

Anthropic has voluntarily partnered with government bodies to assess the risks associated with increasing AI capabilities:

  • Nuclear Information Safety: Collaborating with the Department of Energy’s nuclear experts to assess the risk of models divulging sensitive nuclear weapons information.
  • Capability Testing: Partnering with the U.S. Center for AI Standards and Innovation and the UK AI Security Institute to test models for biological, cyber, and AI R&D capabilities.

Sources

Related