A federal court has granted Anthropic a preliminary injunction blocking the US Department of Defence from designating the AI company as a supply-chain security risk, delivering a significant legal victory that preserves the firm’s ability to compete for government contracts whilst the case proceeds.
The injunction, reported by The Verge and TechCrunch, prevents the Pentagon from implementing a proposed ban that would have prohibited federal agencies and defence contractors from procuring Anthropic’s Claude AI models. The ruling represents an immediate reversal of fortunes for the San Francisco-based company, which faced potential exclusion from the lucrative government AI market.
The dispute centres on the Pentagon’s assessment of Anthropic as a potential supply-chain vulnerability, a designation typically reserved for foreign-owned entities or companies with problematic ownership structures. Anthropic, a US-incorporated firm founded by former OpenAI executives, contested the classification as factually unsupported and procedurally flawed.
According to court filings, the DoD had moved to add Anthropic to its supply-chain risk list without providing the company adequate opportunity to respond to the allegations or present contrary evidence. The preliminary injunction suggests the court found merit in Anthropic’s arguments that the designation process violated administrative law requirements for notice and due process.
The business implications extend beyond Anthropic’s immediate contract pipeline. The case establishes important precedent for how the Pentagon can designate commercial AI providers as security risks, a question of mounting importance as federal agencies accelerate AI procurement. The ruling may embolden other technology firms to challenge similar designations rather than accept exclusion from government work.
For Anthropic, the injunction preserves relationships with existing government customers and maintains eligibility for new contracts whilst the underlying lawsuit proceeds. The company raised $7.3 billion in funding earlier this year, with investors including Google, Salesforce, and Spark Capital. Government contracts, whilst not disclosed in detail, represent a strategic growth area for enterprise-focused AI providers.
The Pentagon faces a more complex landscape. The DoD has sought to balance rapid AI adoption with supply-chain security concerns, particularly regarding Chinese investment in US technology firms. However, overly broad application of security designations risks limiting the pool of capable AI providers available to government agencies at a time when technological superiority is viewed as a national security imperative.
Competitors including OpenAI, which maintains substantial government contracts, and defence-focused AI firms such as Palantir and Scale AI stand to benefit if Anthropic were ultimately excluded from government work. The injunction maintains the status quo, keeping competitive dynamics unchanged for now.
Legal experts note that preliminary injunctions require plaintiffs to demonstrate likelihood of success on the merits, suggesting the court views Anthropic’s legal position favourably. However, the ruling does not resolve the underlying case, which will proceed through discovery and potentially trial.
The case also highlights tensions between administrative efficiency and due process in supply-chain security determinations. Whilst the Pentagon requires authority to act swiftly against genuine threats, companies argue they deserve meaningful opportunity to contest designations that could effectively bar them from an entire market segment.
Industry observers will monitor whether the DoD appeals the injunction or seeks to strengthen its administrative record through supplemental proceedings. The government could potentially refile its designation with additional supporting evidence and enhanced procedural safeguards, though such efforts would face intensified scrutiny given the court’s initial ruling.
The outcome carries implications for broader debates about AI regulation and government procurement. As Congress considers legislation to standardise AI security assessments across federal agencies, the Anthropic case demonstrates the legal vulnerabilities of ad hoc designation processes lacking clear standards and robust procedural protections.
The preliminary injunction represents a critical near-term victory for Anthropic, but the company faces continued uncertainty until the underlying legal questions are resolved. For the Pentagon and the broader government AI market, the ruling signals that supply-chain security designations will face rigorous judicial review when challenged by well-resourced commercial entities.
