Anthropic Nears Deal With US Government to Ease Curbs on Its AI Models

Anthropic, the AI safety-focused company behind the Claude family of models, is close to striking a deal with the US administration to lift restrictions that have curbed the deployment and reach of its AI systems, Bloomberg reported. The agreement, if finalised, would mark a significant regulatory milestone for one of the world's leading AI labs and could reshape how Anthropic's models are accessed internationally.

The precise nature of the current restrictions and the conditions being negotiated were not fully detailed in the reports, but the move reflects a broader US government effort to calibrate AI policy — tightening controls on some fronts while selectively easing them for companies deemed to meet certain safety or compliance standards. Anthropic has long positioned itself as a safety-first organisation, and that reputation may have been a factor in the administration's willingness to negotiate preferential terms.

The development is notable given the near-simultaneous news that OpenAI has restricted its own newest models at government request. Together, the two stories paint a picture of the US government actively managing the AI landscape — deciding not just what gets exported, but which companies earn the right to broader access in the first place.

For markets like India, where Anthropic's Claude models have been gaining traction among enterprise customers and developers, a lifting of restrictions could mean expanded availability and potentially more favourable pricing or partnership terms. The outcome of these negotiations is being closely watched by the global AI industry as a signal of how Washington intends to govern the next generation of AI systems.