US Government Lifts Ban on Anthropic's Mythos 5 Model

The US administration of President Donald Trump has decided to ease restrictions on the use of Mythos 5, an AI model specialized in cybersecurity developed by Anthropic. This powerful technology, which was withdrawn from the market two weeks ago for security reasons, is now being reintroduced for over 100 government agencies and strategically important companies. This decision is seen as a significant step toward enhancing the ability to protect US critical infrastructure from cyberattacks. This was reported by Techcrunch.com reports that

According to Semafor and Reuters, the new directive grants access to the model not only to American companies but also to foreign employees working within these organizations. The previous ban had even barred non-US citizens working at Anthropic itself from accessing the model, which had created serious obstacles for the company's internal workflows and international cooperation.

Security Measures and Restored Trust

In a letter to Tom Brown, Anthropic's Director of Computing, US Commerce Secretary Howard Lutnick emphasized that all necessary security measures for using the model have been taken. According to the Secretary, appropriate protection mechanisms have been established to allow trusted partners to use the Claude Mythos 5 model. However, this permission currently applies only to the Mythos 5 model, while restrictions on the Fable 5 version remain in effect.

As a reminder, the Mythos 5 and Fable 5 models were removed from the market after security researchers discovered that their guardrails could be easily bypassed. The government feared these models could become weapons for launching cyberattacks in the wrong hands. Anthropic specialists have since worked on improving the model's security algorithms.

Anthropic announced on its official X social media page that cooperation with the government has been effective. The company's statement reads: “Since June 12, we have been working closely with the US government to restore access to the Claude Mythos 5 and Fable 5 models. Today, the government informed us that we can provide Mythos 5, our most powerful cybersecurity model, to organizations protecting critical infrastructure”.

The Global Tech Race and Significance for Uzbekistan

This decision by the US government demonstrates an effort to maintain strategic advantage in the global AI race. Cybersecurity models are crucial not only for defense but also for economic stability. For countries undergoing digital transformation, such as Uzbekistan, the regulation of such technologies and their terms of use may define future international cooperation formats.

Anthropic has now begun providing the Mythos 5 model to a limited circle of organizations. Meanwhile, the company continues negotiations with the government to return the Fable 5 model to the general public and further expand access to Mythos 5. This process once again proves how powerful and simultaneously dangerous AI can be.