US Government Orders Anthropic to Restrict Access to New AI Models The sudden shutdown of Anthropic’s newest AI models has turned a technical safety dispute into a geopolitical struggle over who controls frontier AI. In a matter of hours, cutting‑edge tools meant to secure software worldwide became symbols of American power and regulatory chaos.

In early June, Anthropic privately positioned its Mythos 5 system as “too powerful for the public,” citing its skill at finding “high-severity vulnerabilities,” while preparing a safer spin‑off, Fable 5, for general release. Fable 5 launched on June 9 as a “world’s most powerful cybersecurity model,” but within days U.S. officials were alarmed by two developments: access to Mythos through a South Korean telecom they suspected had China ties, and an Amazon research report claiming they could bypass Fable’s guardrails.

On June 12, the Trump administration issued an export‑control order barring use of Mythos 5 and Fable 5 by “any foreign national,” including Anthropic employees abroad and inside the U.S., effectively forcing the company to pull both models entirely. TechCrunch and others report the firm had roughly 90 minutes to comply. The move marked “the first time a government has forced a publicly released frontier artificial-intelligence model to be withdrawn.”

The clampdown immediately disrupted overseas users. A UK entrepreneur described having “the rug pulled from under” his team mid‑project when Fable 5 vanished from their workflow, reinforcing his view that founders must never rely on a single AI tool.

As the weekend unfolded, Anthropic executives and industry allies lobbied Washington, arguing that the vulnerabilities cited in Fable 5 also exist in rival systems and that after‑the‑fact API guardrails are a “brittle interface that can be easily jailbroken,” not a true safety solution. Cybersecurity experts signed an open letter warning that pulling advanced defensive tools from network defenders could itself be “dangerous.”

Globally, the decision is seen as a watershed. The Economist framed it as a warning that Europe must prepare for an era of AI “geopolitical hardening,” while another essay argued that by deciding “who may use the world’s most important technology,” the U.S. has become the gatekeeper to frontier models and most compute. Axios noted that the Fable 5 saga “is turning AI oversight from a theoretical debate into a real test of government power.”

Inside the industry, the episode has sparked soul‑searching about Anthropic’s own messaging. Financial Times and Ars Technica analyses suggest the company talked far more about AI “risk, regulation, or restrictions” than competitors, prompting critics like Meta’s Yann LeCun to say its “fear‑mongering” helped trigger the export ban.

Others see opportunity in decentralization. As one viral post put it, “Fable is banned. Long live local AI,” urging developers to focus on models that can run on personal hardware rather than depend on U.S.‑controlled cloud systems.

Whether the order is eventually softened or upheld, analysts say the outcome could define export‑control rules for future frontier systems and reshape how AI labs balance global markets, safety rhetoric, and national security.

Continue reading https://foxvector.com/stories/019ef059-aaf4-2ed6-73e7-3f47b476bf87