Anthropic has launched a specialized set of AI models designed for U.S. national security customers.
The new models, called Claude Gov, are based on feedback from government agencies. They aim to meet operational needs in areas such as intelligence analysis, cybersecurity, and strategic planning.
In a recent blog post, the company stated that these models are already in use by top-level U.S. security agencies. Therefore, access is restricted to those operating in classified environments.
Claude Gov
Claude Gov is built for secure government use. It exists as a stark contrast to Anthropic’s less guarded consumer and enterprise versions.
These models can work with sensitive data, help plan missions, and support intelligence teams.
According to Anthropic, Claude Gov models handle classified material with better context and fewer rejections, and even understand critical languages and dialects used in national security.
As an upgrade, these models also underwent the same safety testing as other Claude models.
Also read: Anthropic’s AI Found Blackmailing Developers in Simulations
Government AI
Anthropic is expanding into the public sector as it seeks new, steady revenue sources. In November, the company partnered with Palantir and Amazon Web Services (AWS) to promote its AI tools to defense customers.
AWS is also a major investor in Anthropic through Amazon. This collaboration boosts Anthropic’s ability to deliver AI services to agencies that require secure, reliable systems.
Built With Guardrails
Anthropic emphasized that Claude Gov models follow strict safety standards. Despite operating in highly sensitive areas, they are still designed to minimize risk.
They went through the same rigorous testing process as other Claude models. That includes checks for accuracy, reliability, and ethical use.
