Anthropic Pentagon Watch

Pentagon AI Targeting Rules Raise Stakes for Anthropic

Monday, June 29, 2026 · 7 min

Anthropic Pentagon Watch cover art

Anthropic Pentagon tensions now sit inside a wider military-AI shift: revised targeting guidance envisions AI-initiated actions with human monitoring, while court splits over Claude and Mythos approvals show government power shaping frontier-model access.

Listen

Listen to the audio episode

Read the episode transcript

Show notes

Anthropic Pentagon tensions now sit inside a wider military-AI shift: revised targeting guidance envisions AI-initiated actions with human monitoring, while court splits over Claude and Mythos approvals show government power shaping frontier-model access.

In this episode

  1. Pentagon sees broader role for AI in setting military targets | The Straits Times — The Straits Times

    Pentagon sees broader role for AI in setting military targets | The Straits Times myST+ Menu # Pentagon sees broader role for AI in setting military targets Sign up now: Get ST's newsletters delivered to your inbox The US Defence Department has revised targeting guidance as it seeks to speed up AI adoption. PHOTO: AFP Published Jun 26, 2026, 08:50 AM Updated Jun 26, 2026, 10:51 PM Set as…

  2. Step Back — Step back for me: if a judge temporarily blocks the Pentagon’s Anthropic blacklisting, what actually changes in the real world — can agencies and defense contractors go back to using Claude, or can informal national-security warnings still freeze the company out?

    Background sources

  3. The US Government Picked Who Gets Anthropic's Mythos — and It's Not You | Singularity.Kiwi — Singularity.Kiwi

    The US Government Picked Who Gets Anthropic's Mythos — and It's Not You | Singularity.Kiwi News 27 June 2026 · 18:00 NZST ## The US Government Picked Who Gets Anthropic's Mythos — and It's Not You Anthropic can now deploy Mythos — but only to government-approved 'trusted partners.' The policy that started as abstract concern is now picking winners and losers in the AI…