Anthropic promoted India as a priority market, while a U.S. export-control order forced the company to disable its newest models worldwide. The Commerce Department directive named two models, Fable 5 and Mythos 5, and required foreign nationals be cut off from access, a step Anthropic said compelled a global shutdown of those versions. The move arrives just after Anthropic announced a major partnership with Tata Consultancy Services and amid a confidential IPO filing and a large financing round that placed the firm near the frontier tier of AI companies. U.S. officials have said the pause could last a few weeks, and Indian founders, investors and policymakers are already debating whether to accelerate domestic model development or rely more on open-source alternatives.

Anthropic called India one of its largest markets, while Washington told the company specific safety concerns left no room for a partial response.

The Commerce Department order required Anthropic to block access to Fable 5 and Mythos 5 for foreign nationals, the company said. Anthropic told customers it had to disable both models for all users to comply. The firm said it would keep other models available, but the abrupt suspension effectively halted recent product rollouts that had only just reached enterprise and government testers.

The Commerce Department identified a way to bypass safety guardrails in Fable 5 and alleged a narrow "jailbreak" vulnerability in Mythos 5, according to Anthropic's account of the directive. Anthropic disputed the government’s characterization, saying it was presented with verbal evidence of a limited issue and that no universal jailbreak had been found. Even so, the company accepted the need to comply and shut the two models down globally rather than apply a partial block that would have excluded only some users.

The withdrawal covered foreign-national employees and customers outside the United States, Anthropic said, a step it argued was legally necessary to satisfy the order. The models had been launched recently and were being trialled by enterprise and government testers. Anthropic characterised its safeguards as strong and noted that some users had complained those protections were overly broad.

Security concerns and testing

U.S. agencies had been testing Mythos, and officials raised specific worries about the model's potential to assist in identifying software vulnerabilities that could be used in cyber attacks on critical infrastructure, reports said. Anthropic said there was no evidence of a universal vulnerability but nevertheless complied with the order.

Mastheads

U.S. officials said the models could remain locked for a few weeks while national security agencies assess the issue. That timeline is uncertain, and the official cautioned that access might not be restored immediately. Investors and observers now weigh that regulatory and geopolitical risk alongside Anthropic’s commercial plans.

The company has been operating under heightened scrutiny this year. Anthropic has been designated for supply-chain risk by U.S. authorities and has had public disputes over possible military uses of its technology, pressure that adds context to the Commerce Department action and to the company's caution in complying.

India plans and local reaction

Anthropic had described India as one of its largest markets and announced a partnership with Tata Consultancy Services to expand enterprise AI adoption there. The sudden suspension reopened a debate among Indian founders, investors and policy experts about whether the country should speed up development of domestic models, increase investment in open-source alternatives, or remain dependent on a small number of U.S.-based frontier providers.

Aakrit Vaish, founder of Indian AI venture platform Activate, said the decision "completely changes things" and that he woke up "shocked and confused" after the announcement. That reaction is typical of a community that had anticipated deeper commercial ties and local deployments under the Tata Consultancy Services partnership.

For Indian companies that had begun to integrate Anthropic models into products or proofs of concept, the pause raises immediate procurement and contingency questions. For investors, the episode highlights regulatory exposure as a material risk when a handful of U.S. firms supply frontier models used worldwide.

Policy experts in India and elsewhere will be watching whether the suspension prompts concrete steps to build domestic modelling capacity or whether procurement shifts toward open-source systems that governments perceive to be less vulnerable to extraterritorial export controls. The briefings from Washington have already pushed the discussion from hypothetical to urgent.

Anthropic itself is large enough for the episode to matter beyond any single market. The company has filed confidentially for an initial public offering and completed a substantial financing round that placed it near the frontier tier of AI firms. The Commerce Department action arrives against that financial backdrop, complicating investor assessments of regulatory and geopolitical exposure.

For now, Anthropic's immediate technical posture is narrow: other models remain online and the company says it's cooperating with authorities. But the global nature of enterprise AI deployments means a national security finding in Washington can ripple quickly into commercial plans from Mumbai to London.

Related Articles

A U.S. national security assessment, expected in a matter of weeks, will decide whether Fable 5 and Mythos 5 are restored. That ruling will influence whether India accelerates domestic model-building or shifts procurement toward open-source alternatives.

This article was created with AI assistance.