Overview

• Anthropic has deployed Claude Gov, a suite of AI models tailored for U.S. defense and intelligence use, at the highest levels of classified national security operations.
• The models are calibrated to handle classified data with fewer refusal guardrails and offer enhanced proficiency in mission-critical languages and cybersecurity contexts.
• Designed to support tasks such as intelligence analysis, threat assessment, strategic planning and operational support, Claude Gov reflects direct feedback from government customers.
• Anthropic retained its standard safety testing regime but introduced contractual exceptions to align usage policies with specific agency missions.
• The release follows similar defense offerings from OpenAI, Google and Meta as AI firms pursue lucrative government contracts under shifting federal policies.