Anthropic has introduced the Life Sciences Verification Program (LSVP), a beta program that gives verified life science teams and institutions access to its Mythos, Opus, and Sonnet models with safeguards that are more permissive for biology-related work. Dozens of organizations were onboarded through early access, and applications are now open to the broader community. The program currently targets teams and institutions; individual Pro and Max plan access is planned for later.

Applicants must pass a verification process covering research credentials, security standards, and ethical research oversight. Once verified, teams can apply for two grant types. Standard Use grants cover most life science work, including basic science, R&D, supply chain and manufacturing, clinical development, quality assurance, regulatory affairs, and investing and diligence. They can be extended to entire teams, are renewed annually, and apply to Mythos 5.1, Opus 5, Sonnet 5, and future models. High-risk Use grants are an add-on for work blocked under Standard Use, such as specific dual-use research projects. They remove all safeguards blocking life sciences requests, apply to a single research project rather than a full team, and must be renewed every six months. High-risk grants are available today for Claude Opus 5 and Claude Sonnet 5; for Claude Mythos, they remain limited to a small set of entities with additional vetting while Anthropic works with the US government on broader availability. Other safeguards, such as cyber classifiers, remain in place for all LSVP grants.

The program was designed around three threat models: access compromise, insider threats, and agent misuse. Because valid biology research and harmful misuse are often impossible to distinguish at the individual request level, LSVP shifts safeguards from real-time blocking to offline monitoring under a shared-responsibility model. Each entity’s access is tied to the use cases specified in its grant application, and LSVP traffic is continuously monitored against that stated safe scope. Unauthorized activity is flagged to organization admins for triage within pre-agreed timeframes. To support this monitoring, LSVP requires 30-day data retention for flagged activity; this data is compartmentalized, cannot be used for model training, and is not accessible to Anthropic‘s life sciences research teams.

Today LSVP is available in the first-party console for API usage and in Claude for Enterprise and Team plans. It is not yet available on individual plans or third-party platforms, and as a beta it is not available for BAA-enabled organizations, so customers using PHI data should use separate non-BAA orgs. In API and Claude Science, users can switch between grants natively; in Claude.ai and Claude Code, only a preselected default grant applies initially, except when Claude Code is used with API authentication. Xaira, Edison, and Manifold Bio are quoted as early adopters, and Anthropic says it expects to enroll hundreds of organizations within the first week.