The Department of Defense breached its contract with Anthropic and then tried to destroy the company
anthropic
| Source: Mastodon | Original article
The U.S. Department of Defense’s effort to void its $200 million contract with Anthropic has been stopped by a federal judge, who issued an injunction against the government’s actions on grounds of First‑Amendment, due‑process and Administrative Procedure Act violations. Judge Rita Lin’s ruling marks a decisive win for the AI start‑up after the Pentagon, citing “supply‑chain risk,” unilaterally tried to renegotiate and then cancel the agreement that allowed Anthropic’s Claude model to run on classified systems.
The dispute follows a series of escalations that began in February when President Trump ordered agencies to cease using Anthropic’s technology and Defense Secretary Pete Hegseth formally labeled the company a supply‑chain risk. Anthropic’s contract cancellation came just weeks after the firm unveiled its Claude Mythos model—a system we first covered on April 8, 2026, noting its dual‑use potential for both cybersecurity defense and offensive attacks. The Pentagon’s move not only jeopardises Anthropic’s revenue stream but also signals a broader shift in how the U.S. military evaluates and sources AI, potentially favoring rivals such as OpenAI, which has already secured a separate defense deal.
The ruling matters because it sets a legal precedent for how far the government can go in branding a vendor a security threat without breaching contractual and constitutional safeguards. It also raises questions about the future of AI procurement standards, the transparency of supply‑chain risk designations, and the balance between national security and commercial innovation.
Watch for an appeal from the Defense Department, congressional hearings on AI procurement policy, and the Pentagon’s next vendor selection. Industry observers will also be tracking whether Anthropic can resume work on Mythos for civilian and allied‑government customers, and how the decision influences other AI firms navigating the increasingly politicised defense market.
Sources
Back to AIPULSEN