Cybersecurity — 2026-04-17

Bannon publicly sides with Anthropic against Pentagon over autonomous-weapons use of Claude

Steve Bannon, in his War Room broadcast, sided with Anthropic in its multi-month dispute with the Department of War over Pentagon use of Claude, calling fully autonomous lethal AI almost too dangerous and arguing the company's hard limits on autonomous weapons and mass surveillance were defensible. The Pentagon designated Anthropic a supply-chain risk in late February for refusing all lawful uses access. The D.C. Circuit on April 8 declined to stay the designation while a parallel San Francisco district-court injunction bars the ban from being enforced outside DoD. Bannon's intervention pulls a populist-right voice into a fight that had divided largely along industry lines.

Analysis
Bannon's substantive endorsement of Anthropic's autonomous-weapons limits, arguing on the merits that full lethal AI autonomy is too dangerous rather than raising a procedural objection, fractures the assumption that the populist right uniformly backs Hegseth's supply-chain-risk designation and very likely complicates the political coalition the Pentagon assembled for the fight. Bannon alone carries only moderate confidence of moving the litigation; the D.C. Circuit proceeds toward its late-April merits hearing regardless, and the parallel San Francisco injunction continues to bar enforcement outside DoD. The fracture does, however, create oxygen for a settlement in which Anthropic accepts limited DoD use under written safeguards, and whether other right-aligned voices follow within 30 days is the critical observable. If they do, the White House faces rising pressure to signal openness to a negotiated framework rather than litigating to a circuit ruling that sets binding precedent for both parties.
1 sources
  1. Steve Bannon sides with Anthropic in fight with Pentagon: It's almost too dangerous - TheHill

View in full brief →

UNCLASSIFIED // OPEN SOURCE