Anthropic's new Claude Mythos Preview cybersecurity model may help repair strained relations with the Trump administration after months of conflict over surveillance and autonomous weapons.

The Trump administration spent nearly two months publicly attacking Anthropic, calling it a 'RADICAL LEFT, WOKE COMPANY' and a national security threat

Relations deteriorated in late February when Anthropic refused to compromise on two core principles: refusing to enable domestic mass surveillance or develop lethal autonomous weapons without human control

Anthropic's new cybersecurity-focused model, Claude Mythos Preview, may help thaw tensions between the AI company and the Pentagon

The company's ethical stance previously restricted Pentagon use of its technology, creating conflict with the current administration