- Anthropic was reportedly warned Mythos is a super weapon needing gun license-like control
- Within days of release the US government asked Anthropic to take Mythos offline for all foreign nationals
- Anthropic suspended all access, calling the US order a misunderstanding and seeks resolution
Anthropic CEO Dario Amodei reportedly revealed companies testing Mythos warned it was a super weapon and said using it should require a "gun license," according to post on X by Polymarket (which claims to be the world's largest decentralised prediction market that allows users to bet on the outcomes of real-world events using cryptocurrency). NDTV could not independently verify this statement or where Amodei said this.
Fable 5 is built on Anthropic's underlying model Mythos, which the company previewed in April and kept it away from mass release because it was deemed too powerful amid concerns that bad actors could use it to hack critical infrastructure such as banking systems or build bioweapons. According to the company, Mythos was able to identify flaws in every major operating system and web browser it tested, some of these vulnerabilities were reportedly lying undetected for decades. Hence, the company launched a controlled program called Project Glasswing, sharing it with around 50 vetted organisations, including Google, Apple, Amazon, Microsoft, and CrowdStrike, to utilise for defensive cybersecurity work.
Also Read | Opinion | The AI Kill Switch, And Why Sovereign AI Just Became Non-Negotiable
Last Tuesday, the company released a version of Mythos with guardrails to block responses in high-risk areas like cybersecurity and biology. If someone tried to utilise the latest version Fable 5 to hack into a critical system or build a bioweapon, the model would automatically slip back into the earlier less-capable version called Claude Opus 4.8.
According to Vals AI, which tracks AI model performance, Fable 5 ranked as the most capable publicly available AI model on its benchmark tests.
Anthropic has been in active talks to figure out the situation which it calls a "misunderstanding."
The US government on Friday ordered Anthropic to immediately revoke access for foreign nationals (including those employed by Anthropic) to two of its most powerful AI models - Claude Fable 5 and Claude Mythos 5 - citing national security concerns. To comply, Anthropic went ahead and suspended access for all. The directive was to revoke access for all foreign nationals, whether inside the US or outside.
Also Read | Why Anthropic Disabled Its Most Powerful AI Model Days After Launch
Anthropic put out a lengthy statement in which it said: "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance." It added it believes "this is a misunderstanding" and the company is working to restore access "as soon as possible."
The government's "export control directive" to Anthropic came amid concerns of a potential "jailbreak." In the statement it released, Anthropic said, the government has provided only verbal evidence of a "potential narrow, non-universal jailbreak". Simply put, a jailbreak is a technique used to bypass an AI model's built-in safety restrictions and obtain responses it would normally refuse to provide.
