Reports indicate that the US government's order to halt distribution of Claude Mythos 5 and Claude Fable 5 was due to 'suspicions that a group with ties to China had accessed them.'



Regarding Anthropic's latest AI models, ' Claude Mythos 5 ' and ' Claude Fable 5 ,' which have been ordered to be taken off the market by the US government, news outlet Semafor reported that one of the reasons for the order was that 'a group linked to the Chinese government is suspected of having accessed them.'

White House move to limit Anthropic linked to concerns about Chinese access to Mythos | Semafor
https://www.semafor.com/article/06/13/2026/white-house-move-to-limit-anthropic-linked-to-concerns-about-chinese-access-to-mythos



China may have accessed Mythos | The Verge
https://www.theverge.com/ai-artificial-intelligence/949644/china-white-house-anthropic-mythos

On April 7, 2026, AI company Anthropic announced 'Claude Mythos Preview,' a high-performance AI model capable of sophisticated cyberattacks. Claude Mythos Preview boasts an extremely high ability to 'discover and exploit software vulnerabilities,' having discovered 6,202 potential vulnerabilities estimated to be of high severity or higher , and has been shown to be capable of autonomously executing a complete network takeover attack , fully demonstrating its impressive capabilities.

Anthropic develops 'Claude Mythos Preview,' an AI with extremely high cyberattack capabilities, and has also launched 'Project Glasswing,' which will provide a preview version to Microsoft, Apple, and others - GIGAZINE



Subsequently, on June 10, 2026, Anthropic released 'Claude Mythos 5,' the official version of Claude Mythos, and 'Claude Fable,' which is available to all users with enhanced security measures. Claude Mythos 5 and Claude Fable 5 are positioned as updated versions of Claude Mythos Preview, and are touted as having the best performance of any AI model Anthropic has ever released.

The official version of 'Claude Mythos' has finally been released, and 'Claude Fable,' a version with no user restrictions, has also been released, making it available to everyone - GIGAZINE



However, just three days after its release, Anthropic stopped offering Claude Mythos 5 and Claude Fable 5, citing 'directives from the U.S. government's national security authorities.'

Anthropic has stopped distributing 'Claude Fable 5' and 'Claude Mythos 5' at the direction of the US government - GIGAZINE



News outlet Semafor reports, citing sources, that 'the reason the U.S. government ordered Anthropic to stop providing Claude Mythos 5 and Claude Fable 5 is because a group linked to the Chinese government is suspected of having accessed Claude Mythos.'

It remains unclear how the White House obtained information that 'a group linked to the Chinese government had accessed Mythos,' which models the potentially accessed group had access to, and how they obtained access.

Semafor reports that 'if the Chinese government gains access to Mythos, it could pose a national security risk to the United States. Furthermore, Chinese companies may attempt to reverse-engineer and replicate this AI model (Mythos) through a process called distillation .'

David Sachs, an advisor to President Donald Trump, explained on X how the White House ordered Anthropic to stop providing Claude Mythos 5 and Claude Fable 5.




Sachs stated, 'Earlier this week, Anthropic released a Mythos-class model under the commercial name Fable. Fable is Mythos with guardrails. However, if those guardrails fail, Mythos and its advanced cyber capabilities could be exposed to people who shouldn't have them.' 'A trusted partner of both Anthropic and the U.S. government, who was testing Fable, publicly disclosed a jailbreak of those guardrails. Following this, the government asked Anthropic to fix the jailbreak or take the model private.' 'Anthropic claimed the jailbreak was not serious. That contradicts the views of our trusted partner and the U.S. government. It also doesn't seem like appropriate language for Anthropic as an AI security company. I don't think a jailbreak that enables the operation of cyber weapons can be defined as 'not serious'.' 'Therefore, the government issued export restrictions, which Anthropic reluctantly complied with.' 'The U.S. government hopes that Anthropic will rectify the security issues, the export restrictions will be lifted, and Claude Mythos 5 and Claude Fable will be released.' 'The suspension of service for number 5 will be lifted,' he said.

- Continued
The reason Amazon's CEO expressed concerns about 'jailbreaking' the games from Anthropic's Claude Mythos 5/Fable 5 to Trump administration officials was because the company's CEO did so - GIGAZINE



Related Posts:

in AI, Posted by logu_ii