Anthropic’s Mythos 5 AI model returns with restrictions: Here’s who can use it
Anthropic is bringing back its Mythos 5 AI model after receiving partial approval from the US government. Access remains limited to approved organisations.
Published By: Shubham Arora | Published: Jun 27, 2026, 04:14 PM (IST)
Anthropic has received partial approval from the US government to restore access to its Mythos 5 AI model after nearly two weeks of restrictions. The decision means the company's most advanced cybersecurity-focused model can once again be used, but only by a limited group of approved organisations.
The move comes after discussions between Anthropic and the US government over how powerful AI models should be released and who should be allowed to use them. While Mythos 5 is returning for a select group of users, Anthropic's Fable 5 model is still unavailable for general access and there is no official timeline for its return.
Mythos 5 returns with restrictions
According to Anthropic, the company has been informed that Mythos 5 can now be deployed to a group of trusted organisations involved in cybersecurity and critical infrastructure protection. The company said it is working to restore access for these users as quickly as possible while continuing discussions with the government to expand availability.
The approval does not remove the broader restrictions announced earlier this month. Instead, it creates an exception that allows only approved organisations to use the model. Reports suggest the list includes government agencies, trusted enterprises and infrastructure providers, with more than 100 organisations expected to receive access.
The revised arrangement also allows authorised non-US employees of Anthropic and approved organisations to work with Mythos 5 under the updated rules.
Why the model was blocked
The restrictions were first introduced on June 12 after the Trump administration raised concerns about advanced AI models falling into the hands of foreign adversaries. Officials argued that highly capable AI systems could potentially be misused by military or intelligence organisations in countries such as China and Russia.
As a result, Anthropic temporarily withdrew both Mythos 5 and Fable 5 from users while discussions with the government continued. The decision also affected some of the company's own employees, who were unable to access the models because of the licensing rules.
The government has not withdrawn the original order completely. Instead, it has relaxed some of the requirements for Mythos 5 while keeping the remaining restrictions in place until further notice.
Debate over access continues
The decision has also restarted the debate around who should control access to advanced AI systems. The US government has not publicly explained how organisations are selected or what criteria are being used for approval.
The issue has drawn attention across the AI industry. OpenAI CEO Sam Altman recently said safety testing of powerful AI models is important, but questioned whether governments should decide which customers get access.
Anthropic has said it will continue working with US authorities to expand access to Mythos 5 and eventually bring Fable 5 back for general use.
AI companies wait for broader rollout
The latest decision gives Anthropic a path similar to the one followed by OpenAI for its cybersecurity-focused models, where early access is limited to trusted organisations before a wider release.
For Anthropic, this means Mythos 5 is no longer completely unavailable, but it is still far from a public launch. The company is expected to continue working with regulators on future safety protocols and release standards before access is expanded further.
Get latest Tech and Auto news from Techlusive on our WhatsApp Channel, Facebook, X (Twitter), Instagram and YouTube.