The Trump administration has imposed strict export controls on Anthropic's latest AI models, citing national security risks from 'jailbreaking' techniques that bypass safety rules. The restrictions, effective June 12, came days after the release of Claude Fable 5 and Mythos 5.

Anthropic downplayed the risks, calling them minor and exaggerated, noting similar vulnerabilities exist in rival platforms. However, the US Commerce Department effectively banned foreign nationals worldwide, including Anthropic's own staff, from using the models, forcing the company to suspend global access.

The move coincided with Anthropic's plans for a public listing this fall, aiming to raise tens of billions of dollars. The crackdown has drawn sharp criticism from over 170 tech executives, who warned it risks undermining US AI leadership while China advances.

At the G7 summit in Evian, Indian Prime Minister Narendra Modi called for 'broad and inclusive' access to US AI models, while the UK's request for an exemption was rejected. French President Emmanuel Macron described Washington's ability to cut off access as a 'kill switch', urging EU AI sovereignty.

Risto Uuk of the Future of Life Institute called the US move 'hasty and uninformed', advocating for clearer regulations similar to the EU's upcoming rules. German economist Clemens Fuest warned the ban highlights Europe's vulnerability, as the bloc controls less than 5% of global AI infrastructure versus 75% for the US.

Anthropic previously clashed with the Pentagon over refusing to allow its models for mass surveillance or autonomous weapons, leading to contract threats. The company is known for its safety-first 'Constitutional AI' approach.

Despite the controversy, IPO expert Jay Ritter believes the market expects the restrictions to be lifted soon, with an 85% probability of an IPO announcement before November 1. Anthropic CEO Dario Amodei urged G7 leaders to prioritize international cooperation over unilateral actions.

Source: www.dw.com