Anthropic released Claude for Government on October 1, 2026, saying it is now available to US federal and state agencies. It is the company's first civilian agency update since March, when the Pentagon labeled it a supply chain risk.
The platform reported no specific benchmark score, but it runs in a FedRAMP High environment, the strictest US cloud security level. That compares with its previous enterprise deployment, which had less stringent security requirements.
Claude for Government is built on the same architecture as its enterprise version and targets civilian agencies. Availability begins with a public beta, initially for federal and state agencies.
"The platform is designed to meet the highest security standards while providing the same capabilities as our enterprise customers," said Dario Amodei, CEO of Anthropic. The company emphasized that agencies will pay for what they use, with fixed caps.
The announcement follows a recent court ruling that struck down one of two designations against Anthropic as unlawful retaliation. The company did not say whether it will seek to overturn the other designation, and it remains unclear how the Pentagon will respond.
Source: thedecoder
