AI: A gun license should be required to operate this AI! Testers have dubbed it a 'super weapon'—find out why..
In the world of artificial intelligence, Anthropic's new AI model, Mythos, has been consistently making headlines. A new report claims that, upon witnessing the model's capabilities, some testers described it as a "super weapon" and even suggested that its use should require a permit similar to a "gun license."
This claim emerged from a post on the social media platform X by Polymarket. According to the post, Anthropic CEO Dario Amodei stated that some individuals testing the model had raised serious concerns regarding its power. However, this statement has not been independently verified, and it remains unclear where Amodei made these remarks.
What exactly is Mythos?
Mythos is the underlying technology behind Anthropic's new AI model, Fable 5. The company offered a glimpse of it in April but did not release it to the general public, citing its extraordinary capabilities and associated safety concerns.
The company believed that if this technology fell into the wrong hands, it could be used for activities such as hacking critical infrastructure—like banking systems—or developing biological weapons.
According to Anthropic, during testing, Mythos discovered security vulnerabilities in almost every major operating system and web browser. Some of these flaws had reportedly remained undetected for decades.
Restricted to select companies
In light of these risks, the company launched a controlled initiative called "Project Glasswing." Under this program, Mythos was shared with only about 50 trusted entities, including major companies like Google, Apple, Amazon, Microsoft, and CrowdStrike. The objective was to conduct defensive research aimed at strengthening cybersecurity.
Last week, Anthropic released a new version of Mythos that incorporates several safety restrictions. According to the company, if a user attempts to employ Fable 5 for high-risk activities—such as hacking critical systems or creating biological weapons—the model will automatically revert to the older, less capable version, Claude Opus 4.8.
Vals AI, an organization that monitors AI model capabilities, ranks Fable 5 as one of the most capable AI models currently available to the public, noting that it has topped benchmark tests.
**US Government Imposes Immediate Ban**
The situation escalated when the US government ordered Anthropic to immediately revoke foreign nationals' access to its two most powerful models: Claude Fable 5 and Claude Mythos 5. The government cited national security concerns as the reason for this directive.
In compliance with the order, Anthropic temporarily suspended access for all foreign users. The mandate applied not only to individuals residing outside the US but also to foreign nationals currently within the country.
According to reports, the government fears these models could be susceptible to "jailbreaking." In AI terminology, a jailbreak refers to a technique used to bypass a model's safety guardrails and elicit responses it would normally refuse to provide. Consequently, Anthropic's most powerful AI models have become the subject of intense debate, centering not only on their technical capabilities but also on issues of safety and national interest.
Disclaimer: This content has been sourced and edited from Amar Ujala. While we have made modifications for clarity and presentation, the original content belongs to its respective authors and website. We do not claim ownership of the content.
