Anthropic’s new cybersecurity model could get it back in the government’s good graces

Source: The Verge

Background

The Trump administration spent nearly two months in a public dispute with AI company Anthropic, labeling the firm a “radical left, woke company” full of “left‑wing nut jobs” and calling it a menace to national security.

New Cybersecurity Model

Anthropic’s relationship with the Pentagon began to thaw after the company announced its new cybersecurity‑focused model, Claude Mythos Preview. The model is designed to address security concerns and demonstrate Anthropic’s commitment to responsible AI use.

Potential Reconciliation

The Pentagon’s concerns originally stemmed from two red lines Anthropic would not cross:

• Using its technology for domestic mass surveillance.
• Deploying lethal fully autonomous weapons without human oversight.

By introducing Claude Mythos Preview, Anthropic appears to be taking steps that could satisfy the administration’s security requirements and improve its standing with the government.

Read the full story at The Verge.