Skip to playerSkip to main content
FABLE 5 AI BANNED BY US GOVERNMENT AFTER JUST 3 DAYS – ANTHROPIC FIGHTS BACK
https://www.newsplusglobe.com/
Anthropic released Claude Fable 5 on June 9, 2026. By June 12, the US government ordered the company to block all access to Fable 5 and its more powerful sibling Mythos 5 for any foreign national – including foreign national employees at Anthropic. The government cited a potential "jailbreak" that could identify cybersecurity vulnerabilities. But Anthropic is calling it a "misunderstanding" and working to restore access.

The company warns that if this standard is applied across the industry, it would "essentially halt all new model deployments for all frontier model providers." Pre-launch testing ran for thousands of hours with the US government, UK AISI, and multiple third-party organizations – and no tester found a universal jailbreak. Anthropic also notes that the capability cited is "widely available from other models including OpenAI's GPT-5.5."

Should governments have the power to ban AI models mid-deployment? Drop your take in the comments.

#Fable5 #Anthropic #AI #BreakingNews #NationalSecurity #AIRegulation

Category

🗞
News
Transcript
00:00Stop scrolling! An AI model called Fable 5 was released on Tuesday. By Friday, the U.S. government had banned
00:08it. Here is what happened.
00:10Fable 5 is Anthropics' most powerful AI model yet. The company called it the most powerful model we have ever
00:16made generally available.
00:18It can code for days, solve complex scientific problems, and handle tasks no previous model could sustain.
00:24On June 12, the U.S. government issued an emergency directive. It requires Anthropics to block all access to Fable
00:325 by any foreign national, including foreign national employees at Anthropics.
00:36The government cited a potential jailbreak that could allow cyber attacks on critical infrastructure. But here is the twist. Anthropics
00:44is fighting back.
00:45They called it a misunderstanding and are working to restore access. They say the capability the government is worried about
00:52is already available in other models, including OpenAI's GPT 5.5, which is used every day by cybersecurity defenders.
01:00Anthropics warns that if the government applies this standard across the industry, it would essentially halt all new model deployments
01:07for all frontier model providers.
01:09They also revealed that pre-launch testing ran for thousands of hours with the U.S. government, and no tester
01:15found a universal jailbreak.
01:17Comment below, should the government have the power to ban an AI model?
Comments

Recommended