On July 9, 2025, Lawrence Livermore National Laboratory (LLNL) announced it is extending its deployment of Anthropic’s Claude for Enterprise to cover the whole laboratory. The rollout will make advanced AI tools available to roughly 10,000 scientists, researchers and staff. According to LLNL, this represents one of the largest Claude for Enterprise deployments within the U.S. Department of Energy’s national laboratory system.
Why this matters
LLNL conducts research across fields including nuclear deterrence, energy, materials science and energy security. Making Claude broadly available is intended to help researchers process large, complex datasets, generate and test hypotheses more rapidly, and explore new research directions with an AI that is attuned to scientific context. LLNL says the expansion will strengthen research capabilities and support national-security-related missions.
Partnership and leadership comments
LLNL framed the expanded collaboration with Anthropic as an example of how AI can be integrated into government research operations. Thiyagu Ramasamy, Anthropic’s Head of Public Sector, said, “We’re honored to support LLNL’s mission of making the world a safer place through science and technology.” Greg Herweg, Chief Technology Officer at Lawrence Livermore National Laboratory, added that LLNL has long been at the forefront of computational science and that the expanded partnership shows how frontier AI can amplify researchers’ capabilities on pressing global challenges.
Claude features and enterprise security
The Claude application suite used by LLNL includes security features tailored for government environments. The platform’s expanded context window can handle hundreds of documents, entire codebases with more than 100,000 lines, or complex datasets in a single query — enabling comprehensive analysis of tasks such as fusion experiments or nuclear simulations. Enterprise security provisions listed by LLNL include single sign-on (SSO), audit logging, role-based access controls and end-to-end encryption.
Areas of application at LLNL
LLNL scientists are deploying Claude across multiple disciplines, including:
- Emergency response: analyzing data from the National Atmospheric Release Advisory Center (NARAC) to respond to nuclear, radiological, chemical or biological incidents.
- Energy security: advancing fusion energy research building on LLNL’s 2022 achievement of fusion ignition.
- Advanced manufacturing: speeding materials discovery and optimization through AI-driven analysis of 3D printing processes and manufacturing data.
- Computational biology: processing large simulation datasets to support biosecurity research and accelerate detection of biological threats.
- High-performance computing: optimizing code development and scientific computing workflows to better leverage LLNL’s supercomputing resources.
Pilot program and uptake
The full deployment follows a successful pilot program that included the first-ever AI Jam with U.S. national labs and the aiEDGE for Innovation Day in March 2025. About 3,200 LLNL scientists and operational staff took part in those events, getting hands-on experience with how Claude can accelerate and enhance scientific and national-security research.
How to get started
The announcement notes that organizations interested in adopting Claude for Enterprise can contact Anthropic’s public sector team to learn more and begin deployment.
The blog post was updated on July 11, 2025 with additional information from LLNL.
