Federal Judge Says Government Record Worsened in Anthropic Supply Chain Risk Dispute as Court Probes IC Deployment of Mythos AI
U.S. District Judge Rita Lin said Thursday the government's record supporting Anthropic's "supply chain risk" designation "has gotten worse," finding no evidence in the record that Anthropic could alter or disable delivered AI models via a "kill switch"
DOD's near-simultaneous wind-down of Anthropic use and expanded, undisclosed deployment of Mythos in national security applications undercuts the government's litigation posture. Lin's questioning signals she views continued Mythos use as inconsistent with the supply-chain risk designation, strengthening the evidentiary basis for a summary judgment ruling against the government, with moderate confidence given consistent judicial signaling across two hearings though no ruling has yet issued. Bloomberg, Axios, and TechCrunch filed independent primary accounts of the hearing, with FedScoop and Courthouse News Service corroborating without shared-wire dependency. The wind-down may instead reflect a genuine agency shift toward alternative vendors unrelated to the litigation. A ruling for Anthropic would force DOD to abandon the September 30 wind-down and justify continued Mythos deployment under the same trust rationale used to bar Claude; a ruling for the government lets agencies proceed with divestment as the OneGov pilot lapses in late August.
5 sources
- Judge Voices Doubt US Has Justified Its Ban on Anthropic AI -
Bloomberg - Judge says government's case got "worse" in Anthropic fight -
Axios - Judge says Trump admin still lacks evidence for Anthropic 'supply-chain risk' label -
TechCrunch - Judge likely to rid Anthropic of Pentagon's 'supply chain risk' label -
Courthouse News Service - Record has gotten worse for the government in Anthropic dispute judge says -
FedScoop