Silicon Valley executives and global policymakers are grappling with the implications of the Trump administration’s abrupt decision to impose strict export controls on Anthropic’s cutting-edge AI models. The move, enacted on June 12, just days after the release of the company’s latest upgrades—Claude Fable 5 and the more advanced Mythos 5—effectively bans foreign nationals worldwide from accessing these powerful tools. Citing national security risks associated with ‘jailbreaking,’ a technique that bypasses AI safety protocols, Washington’s action has forced San Francisco-based Anthropic to suspend global access entirely. The company, however, has characterized these security concerns as minor and present across rival AI platforms.
Experts and Allies Stunned by US ‘Kill Switch’
The export ban has drawn sharp criticism from the technology sector and international leaders. In an open letter, over 170 tech executives warned that the curbs could jeopardize America’s AI leadership by limiting access to crucial tools for cybersecurity defenders while China’s capabilities continue to advance. At the recent G7 summit in Evian, France, Indian Prime Minister Narendra Modi advocated for ‘broad and inclusive’ access to US AI models. The UK’s request for an exemption from the ban was reportedly rejected.
Some European lawmakers have voiced alarm over Washington’s ability to unilaterally shut down access, dubbing it a ‘kill switch.’ This has amplified calls within the European Union to secure its own AI sovereignty. French President Emmanuel Macron expressed concern, stating, ‘We won’t buy any models made by these companies if overnight, you can just flip the switch.’
Ad Hoc Moves Expose US Regulatory Gaps
Risto Uuk, head of European policy and research at the Future of Life Institute, described the US action as ‘hasty and uninformed.’ He urged Washington to establish clearer AI regulations, drawing a parallel to the EU’s forthcoming AI regulations set to roll out in August. Uuk emphasized that AI safety should not hinge on the ‘goodwill on a given week’ of a single firm.
Traditionally, the US has employed legal tools like export controls against foreign adversaries such as China and Russia. However, targeting an American company like Anthropic sets a potentially damaging precedent, according to tech experts and policymakers. This action risks undermining investor confidence in the AI boom, stifling innovation, and weakening the US’s overall technological edge.
Clemens Fuest, president of Germany’s ifo Institute, highlighted Europe’s ‘vulnerability’ in AI capabilities in a research note. He called for increased investment in data centers, chip factories, and energy infrastructure within the bloc. Despite being a significant AI user, Europe controls less than 5% of global AI infrastructure, compared to the US’s 75% and China’s 15%, according to recent data.
Anthropic’s Pentagon Friction
Earlier this year, Anthropic encountered friction with the US Department of Defense. The company had refused to lift restrictions on its AI models concerning mass surveillance of US citizens or fully autonomous lethal weapons systems. The Pentagon designated Anthropic a ‘supply chain risk’ and threatened to cancel contracts potentially worth hundreds of millions of dollars, leading Anthropic to initiate legal action.
Anthropic has cultivated a reputation as one of the more cautious players in the frontier AI race, prioritizing safety through its ‘Constitutional AI’ approach, where models are trained to self-critique against a set of written principles. However, some, like AI researcher Pedro Domingos, argue that Anthropic itself is exhibiting ‘dangerous overreach’ by positioning itself as a moral authority and usurping governmental functions.
IPO Plans Face Headwinds
The recent controversies arrive at a critical juncture for Anthropic, which recently filed plans for a public listing, likely this fall, aiming to raise tens of billions of dollars. Bankers anticipate the company could raise between $30 billion and $60 billion, positioning it among the largest IPOs in history. The firm was recently valued at nearly a trillion dollars and is reportedly generating revenue estimated at close to $47 billion annually.
However, being targeted by the White House, facing export bans on its flagship models, and losing out on major government contracts are likely to impact investor sentiment. While Wall Street is hopeful that Anthropic, alongside a potential listing for OpenAI, can replicate the enthusiasm seen with SpaceX’s recent $75 billion IPO, the current situation presents challenges.
According to IPO expert Jay Ritter, the market appears to believe that the foreign-access curbs on Anthropic’s top-tier AI models will be lifted relatively soon. He cited the Kalshi prediction platform, which indicates an 85% chance of Anthropic announcing an IPO before November 1, a figure that has seen minimal change in the past week. Ritter noted, ‘There is still enormous enthusiasm in both public and private markets for AI companies.’ In a statement at the G7 summit, Anthropic CEO Dario Amodei urged leaders to prioritize international cooperation on AI regulation over unilateral actions, advising them to ‘resist the temptation to splinter.’
