Technology & Science
US Emergency Export Controls Force Anthropic to Pull Fable 5 & Mythos 5 Worldwide
On 12 June 2026 the US Commerce Department invoked national-security export powers to bar all foreign nationals from Anthropic’s new Fable 5 and Mythos 5 models, leaving the firm no choice but to switch the services off globally within 90 minutes.
Focusing Facts
- Commerce Secretary Howard Lutnick’s directive, issued Friday 12 June 2026 at 1:15 p.m., threatened civil and criminal penalties unless Anthropic blocked the models for every non-US citizen within 90 minutes.
- The order applies even to Anthropic’s own non-US staff and outside adviser Rishi Sunak; lacking citizenship data the company disabled both models for every user worldwide at 5:21 p.m.
- On 15 June 2026 Anthropic flew senior engineers to Washington for face-to-face talks aiming to lift the ban and salvage a planned >$900 billion IPO.
Context
Washington’s snap clamp-down echoes the 1946 McMahon Act’s restriction of nuclear know-how and the 1981–1990 COCOM embargoes on advanced chips: moments when frontier technologies abruptly shifted from commercial commodity to tightly rationed strategic asset. The move signals an accelerating trend—‘AI nationalism’—in which model weights are treated like fissile material, subject to export licences, insider-threat vetting and great-power bargaining. Over the last decade the US had championed open-source AI and global cloud access; this reversal suggests the post-Cold-War ideal of friction-free tech diffusion is ending. If the precedent sticks, future dominant models may be gated by passports, driving parallel ecosystems (US, China, non-aligned) and altering innovation geography for generations. On a 100-year arc, the incident may mark the point when artificial-intelligence governance left voluntary norms and entered the hard-law, state-secrecy domain—much as aviation and nuclear industries did a century earlier—reshaping who can build, study, and benefit from the most powerful cognitive tools humanity creates.
Perspectives
Global financial and business media
Bloomberg, Yahoo! Finance, Economic Times — Interpret the shutdown as a watershed moment where Washington is willing to wield export-control power over AI to guard national security, serving as a stark warning to Silicon Valley investors and firms. Coverage centers on market impact and policy risk—reflecting investor and government sources—so it tends to accept the security rationale at face value while giving limited attention to civil-liberties or labour concerns.
Technology-industry friendly outlets
Republic World, The Times of India, TimesNow — Cast the directive as an excessive, hurried overreach that forced Anthropic to pull its models worldwide, jeopardising innovation and illustrating budding 'AI nationalism'. Reliant on access to tech companies and sympathetic to start-up narratives, these reports foreground Anthropic’s talking points and play down the possibility that real security flaws existed.
Socialist/anti-capitalist press
World Socialist Web Site — Frames the ban as evidence of the capitalist state fusing with big tech to weaponise AI for imperialist war while suppressing public access and democratic control. Ideological commitment to anti-imperialist, anti-capitalist analysis leads to sweeping claims about US militarism and class struggle that may discount technical safety issues raised by officials.
Like what you're reading?