Open claude.ai and it reads, “Claude Fable 5 is currently unavailable.” Click on ‘Know More’ and the website redirects to its maker Anthropic’s statement explaining the US government’s directive to suspend access to the AI model.
Anthropic, the US company behind Claude, on Saturday switched off its two most advanced artificial intelligence models — Fable 5 and Mythos 5 — for every user worldwide, including in India, after the Donald Trump administration ordered it to deny access to all foreign nationals on national security grounds.
India, by Anthropic’s own admission, accounts for the most number of Claude users (5.8% of its global users) after the US (22%).
The company said it received the export control directive on Friday at 5.21pm ET and disabled both models within hours. Because it cannot reliably separate foreign nationals from other users in real time, it shut the models for everyone — Americans included. Access to all of Anthropic’s other models is unaffected.
Anthropic reaction & US response
The directive, a US official confirmed, came from the Commerce Department. Axios reported the letter came from Commerce Secretary Howard Lutnick.
The letter did not spell out the specific concern, Anthropic said, adding that its “understanding is that the government believes it has become aware of a method of bypassing, or ‘jailbreaking’ Fable 5” — a workaround that could let the model be used to find software vulnerabilities, or access other illegal information such as weapons or explosives making.
Fable 5, released just days earlier, is a locked-down public version of Mythos 5, a model Anthropic has held back over its unusually strong ability to spot holes in code that hackers could exploit. Mythos 5 has gone only to a small set of companies and governments as part of what Anthropic calls Project Glasswings. The goal is to allow these selected groups and entities to discover other potential dangerous uses or vulnerabilities in their own infrastructure.
Anthropic pushed back on the US government’s directive and said it had reviewed the technique at the centre of the speculation and found it surfaced only a handful of minor, already-known vulnerabilities that other publicly available models can find too. No safety tester had located a “universal jailbreak”, it said.
The government, it alleged, had offered only “verbal evidence of a potential narrow, non-universal jailbreak”.
“We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people,” the company said.
It warned: “If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers.”
Anthropic said it believed there had been a “misunderstanding” and was working to restore access.
The government’s chief information officer, Kirsten Davies, wrote on X that the defence department backed prioritising national security: “Some things are simply more important than revenue cycles, clickbait, and pre-IPO valuation. America First. Always.”
Also read: The AI price war is here, piling pressure on OpenAI and Anthropic
What the suspension could mean
For years, US export controls have targeted the chips and tools that power AI. Restricting foreign access to a finished model is a sharp escalation — Washington reaching past the hardware to the software itself.
Dean Ball, a former senior AI adviser in the Trump White House and the lead drafter of the administration’s 2025 AI Action Plan, said on X that the order implies all non-Americans, including those based in the US, would be barred: “This means you should expect to have to prove your citizenship to use Anthropic models.”
The directive’s wording also appears to cover Anthropic’s own foreign-born staff.
Reuters named co-founder Chris Olah, AI researcher Andrej Karpathy and philosopher Amanda Askell as having been born abroad, but said it could not determine their citizenship status, and Anthropic declined to comment on whether such employees would lose access.
The move also landed as Anthropic, valued in February at $380 billion, heads towards an initial public offering (IPO).
For India, the episode is a pointed reminder of a familiar vulnerability: the country’s continued reliance on frontier AI it neither builds nor controls. As with past dependencies on foreign technology, an Indian user, startup or institution that might have begun building on Fable 5 this week would have found the tool gone overnight.
Earlier this week, India’s Tata Consultancy Services Ltd (TCS) announced a partnership with Anthropic to help customers scale enterprise AI adoption. TCS had said it will set up a dedicated business unit focused on delivering strong customer value propositions, joint industry solutions and deep AI expertise through early access to Claude models.
To be sure, a number of other Claude models remain available for use.
Also read: OpenAI files for IPO: Why AI giants are racing to Wall Street
Anthropic vs Pentagon
The shut-off is the latest turn in an Anthropic-US government feud that has been building for months. Its origins, and its stakes, were laid out in unusual detail by Ball, the former adviser, on the New York Times podcast The Ezra Klein Show in March.
By Ball’s account, the company’s relationship with the US government dates to the summer of 2024, when the Joe Biden-era Pentagon and Anthropic agreed to use Claude in classified settings — including intelligence analysis — with two carve-outs: no domestic mass surveillance, and no fully autonomous lethal weapons.
The Trump administration expanded the contract in 2025 on the same terms, Ball said, before the relationship soured through the autumn and broke into the open early in 2026. The decisive rupture, he said, came when Emil Michael, the under secretary of war for research and engineering, asked Anthropic to delete the clause barring use of Claude to analyse bulk-collected commercial data — the capability at the heart of the mass-surveillance objection.
The fight was also unusually personal and political. Defence secretary Pete Hegseth wrote on X that Anthropic’s “true objective is unmistakable. To seize veto power over the operational decisions of the United States military. That is unacceptable.”
Trump also branded Anthropic a “radical left, woke company” and its staff as “Left-wing nut jobs”, while Michael called Anthropic chief executive Dario Amodei a “liar” with a “God complex”. Ball also recounted on the podcast that rival entrepreneur Elon Musk — who runs a competing AI firm — had attacked Anthropic relentlessly on X.
Matters came to a head in February this year, when Hegseth threatened to invoke the Defense Production Act and to designate Anthropic a “supply chain risk”.
That label is normally reserved for technology judged too dangerous to sit anywhere in the US military’s supply chain — it has been used against foreign firms such as China’s Huawei over espionage fears — and had never before been turned on an American company.
Trump then announced on Truth Social that every federal agency would stop using Anthropic technology: “We don’t need it, we don’t want it, and will not do business with them again!” The US President warned the company to “be helpful during this phase out period, or I will use the Full Power of the Presidency to make them comply, with major civil and criminal consequences to follow”.
Anthropic, whose Pentagon work fell under a contract worth $200 million, said being designated a supply chain risk “would both be legally unsound and set a dangerous precedent”, and added: “No amount of intimidation or punishment from the Department of War will change our position on mass domestic surveillance or fully autonomous weapons”
The defence department ultimately signed a deal with OpenAI, which says it holds the same red lines as Anthropic. OpenAI chief Sam Altman publicly backed his rival, telling staff in a note seen by the BBC that his company would reject military uses “such as domestic surveillance and autonomous offensive weapons”.
Anthropic subsequently filed two federal lawsuits against the defence department on 9 March, alleging that the ‘supply chain risk’ label was applied improperly and amounts to retaliation for the company’s protected speech on AI safety, in violation of its due process rights.
“The Constitution does not allow the government to wield its enormous power to punish a company for its protected speech. No federal statute authorises the actions taken here,” Anthropic argued, telling a court the designation would cause “irreparable harm” and put “hundreds of millions, or even multiple billions, of dollars in lost revenue” at risk.
The cases remain unresolved: a San Francisco federal judge granted a preliminary injunction barring enforcement of the ban on Claude, but the DC appeals court denied Anthropic an emergency stay on 8 April.
(With inputs from agencies)
