AI Risk Clock
Doomsday Clock5 min to midnight
→
⚖️ Neutral edit
thenextweb.com2026-04-18

Anthropic's Amodei meets Wiles and Bessent at the White House over Mythos access and Pentagon standoff - The Next Web

PolicySafetyIndustry

Anthropic CEO Dario Amodei met with White House Chief of Staff Susie Wiles and Treasury Secretary Scott Bessent on Friday to discuss government access to Mythos, Anthropic's frontier AI model capable of finding thousands of zero-day vulnerabilities. The White House described the talks as 'productive and constructive,' signaling a potential thaw in the standoff that began when the Pentagon blacklisted Anthropic for refusing to drop safety restrictions on the model.

The meeting focused on negotiating access to Mythos, though any agreement would likely exclude the Defence Department and route access through civilian agencies instead. This approach attempts to circumvent the impasse created by Anthropic's safety restrictions, which the Pentagon had required the company to remove for military use.

The discussions represent an effort to balance national security interests with Anthropic's safety protocols, potentially establishing a framework for controlled government use of advanced AI capabilities outside direct military applications. The outcome could set precedents for how frontier AI models are integrated into government operations while addressing developer concerns about responsible deployment.

Read this story in another voice
● REC · 2026