Anthropic, the company that created the Claude artificial intelligence platform, announced the global suspension of its most advanced artificial intelligence models, Fable 5 and Mythos 5.
This decision was taken yesterday, June 12, 2026, after receiving an export control directive issued by the United States government due to alleged national security reasons. The measure requires restricting access to these systems to any foreign citizen, even if they are within US territory or work for the company itself.
Given the impossibility of applying the restriction selectively, Anthropic chose to disable both models for all users worldwide. The company clarified that access to the rest of its artificial intelligence systems remains unchanged.
As the company explained in the official statement, the government’s decision would be related to reports about a possible method to bypass Fable 5’s security safeguards using techniques known as jailbreaksthat is, methods designed to trick the model into ignoring or evading its security restrictions.
It is worth noting that the suspension It occurs only 4 days after the launch of the affected models. On June 9, Anthropic presented Fable 5 as its most advanced system for general use and Mythos 5 as a version with superior capabilities aimed at selected partners through Project Glasswing, a restricted access program to strengthen the defense of critical infrastructure.
The debate revolves around the application of the so-called rules of “deemed export” (considered export), which establish that Sharing sensitive technology with a foreign person is legally equivalent to exporting it to your country of origin.
In response to the accusation, Anthropic maintains that the evidence presented by the authorities is limited to the identification of a small number of previously known and low-complexity software vulnerabilities.
The company also states that other publicly available artificial intelligence models are also capable of detecting these same vulnerabilities without the need to violate protection mechanisms. Furthermore, it ensures that has not received evidence of the existence of a jairbreak universal capable of broadly overriding the security restrictions built into Fable 5.
“Sudden and excessive” restrictions
The measure caused a quick reaction within the technology industry. Dean Ball, a researcher at the Foundation for American Innovation, called the decision “baffling” and questioned whether the United States would restrict the use of its best models to allies while maintaining a more flexible policy in other technological areas.
For his part, Chris McGuire, a specialist at the Council on Foreign Relations, stated that export controls can be a useful tool, but warned that such broad and sudden restrictions could end up harming US leadership in artificial intelligence.
Criticism also arose from the business sector. Dan Shipper, CEO of Every, an AI startup, noted that the outage could significantly impact companies and developers that had already integrated Fable 5 into their operations. In addition, multiple users They expressed frustration on social networks because they paid premium subscriptions specifically to access these models.
Beyond the immediate impact on Anthropic users, the episode marks a potential turning point for the artificial intelligence industry. Until now, U.S. export controls have focused primarily on chips and computing capacity; This decision shows that the models themselves can become the subject of regulatory restrictions.
If this trend consolidates, companies, governments and developers could accelerate the search for local alternatives, open models and technological sovereignty strategies to reduce dependence on US suppliers.
Bittensor (TAO) cryptocurrency price rises
An example of this is that the price of the bittensor cryptocurrency (TAO) has risen more than 15% since the restrictions on Claude became known:
And why does this rise occur? Bittensor’s core value proposition is decentralized, permissionless AI, as an alternative to centralized providers (Anthropic, OpenAI, etc.).
The suspension of Fable 5 and Mythos 5 by government order could be read in the crypto market as a validation of that thesis: Centralized AI exposed to regulatory/political risk and it can be cut from one day to the next, regardless of its technical level or how many users depend on it.
Under that logic, Speculative capital would rotate into networks that (in theory) do not depend on the approval of a single government to continue operatingand Bittensor, as the largest decentralized AI network by capitalization, would be the asset «proxy» most obvious within the «AI tokens» sector to express that bet.