Superpower Daily
Judge Vacates Pentagon’s Anthropic Blacklist and Orders Defense...Databricks Files $5B Equity Offering With Nearly All of It Alre...Snowflake Open-Sources Semi-Persistence With Sub-Second Single-...Cisco Rolls Out MyAgent to 90,000 Employees for Supervised Cros...Midjourney Opens V8.2 Image-Editing Tests With Up to Four Refer...Gemini Omni 1.1 Flash Adds 40-Second Scene Extensions and 4K Fi...Judge Vacates Pentagon’s Anthropic Blacklist and Orders Defense...Databricks Files $5B Equity Offering With Nearly All of It Alre...Snowflake Open-Sources Semi-Persistence With Sub-Second Single-...Cisco Rolls Out MyAgent to 90,000 Employees for Supervised Cros...Midjourney Opens V8.2 Image-Editing Tests With Up to Four Refer...Gemini Omni 1.1 Flash Adds 40-Second Scene Extensions and 4K Fi...
Policypublished

Judge Vacates Pentagon’s Anthropic Blacklist and Orders Defense Department Retraction

The decision removes a department-wide bar on Anthropic technology, while a related Washington case and possible agency-specific restrictions leave the full procurement picture unsettled.

By 3 min read
Judge Vacates Pentagon’s Anthropic Blacklist and Orders Defense Department Retraction
Sign in to save and react

Listen to this story

The audio brief

About 1:38
0:001:38
Read transcript
A federal judge has ordered the Pentagon to withdraw its blacklist of Anthropic, removing a department-wide bar on the company’s technology in Defense Department work. Judge Rita Lin ruled that the Pentagon violated Anthropic’s First Amendment rights and Fifth Amendment due-process protections, and also broke the statutory framework it relied on. She called the designation arbitrary and capricious, and directed the department to retract its related guidance, directives, communications, and instructions. The dispute was not about Anthropic refusing to work with the government. Claude was already being used for intelligence analysis, operational planning, modeling, simulation, and cyber operations. The fight was over contract limits on mass domestic surveillance and weapons that can select and engage targets without human involvement. Anthropic would not remove those restrictions, and negotiations broke down. In March, the Pentagon escalated that supplier dispute into a department-wide supply-chain-risk designation, barring defense contractors from using Anthropic technology in Defense Department work. The judge found that officials also relied on Anthropic’s criticism of the administration’s AI policies and what they described as a hostile public posture. Her ruling says those criticisms could not lawfully justify such sweeping penalties. But the procurement picture is not fully settled. A related case in Washington, D.C., involves a separate designation, and individual federal agencies may still impose their own procurement or security restrictions. The key question now is how those parallel proceedings shape Anthropic’s access to government work.

Story brief

3 key points

The decision narrows the Pentagon’s ability to punish Anthropic over contract disagreements involving Claude’s military use. Judge Rita Lin found that the Department of Defense violated Anthropic’s First Amendment and due-process rights, and also breached the governing statutory framework. The department must withdraw its March designation and related instructions, but the ruling does not resolve separate...

  1. 01

    The dispute centered on Anthropic’s limits covering mass domestic surveillance and autonomous weapons, not a refusal to support government work.

  2. 02

    Claude was already used for intelligence analysis, planning, modeling, simulation, and cyber operations before the designation.

  3. 03

    Lin found the Pentagon’s actions arbitrary and capricious, as well as unconstitutional and procedurally defective.

A federal judge has removed the Pentagon’s supply-chain-risk designation of Anthropic, undoing a measure that barred defense contractors from using the company’s technology in Department of Defense work. The ruling places constitutional limits on the department’s response to a dispute over how Claude could be used.

U.S. District Judge Rita Lin held that the Department of Defense violated Anthropic’s First Amendment rights and Fifth Amendment due-process protections. She directed the department to rescind its guidance, directives, communications, and instructions against the company.

A supplier dispute became a department-wide exclusion

The Pentagon imposed the designation in March after negotiations over military use of Anthropic’s Claude models broke down. The conflict followed Anthropic’s refusal to remove contractual limits on mass domestic surveillance and weapons that select and engage targets without human involvement.

Existing government use complicates the Pentagon’s case

Claude had already been used for intelligence analysis, operational planning, modeling, simulation, and cyber operations. The disagreement was therefore over permissible uses and contract terms, rather than whether Anthropic technology had any government role.

Lin wrote that the defendants cited Anthropic’s hostile public posture and criticism of the administration’s AI views as reasons they could not trust its models’ integrity. She concluded that neither the Constitution nor the statute the department invoked allowed sweeping penalties principally based on the company’s critique of administration views.

The remaining boundary is narrower, not yet settled

Lin had already issued a preliminary injunction on March 26, finding Anthropic likely to prevail on First Amendment retaliation, Fifth Amendment due process, and procedural claims. At the merits stage, she also found the designation violated the governing statutory framework and was arbitrary and capricious.

Anthropic’s related case in Washington, D.C., remains pending because the Pentagon used two distinct designations that must be litigated in separate courts. This ruling also does not establish whether individual federal agencies retain separate procurement or security restrictions.

Editorial analysis

Our Read

The important split is between procurement discretion and a government-wide security designation. The court removed the latter after finding constitutional and legal defects; it did not answer every question about narrower agency decisions. That distinction will matter to AI vendors that sell to government while keeping limits on sensitive uses. The next concrete tests are whether agencies retain or create separate restrictions and how the related Washington, D.C., case resolves the Pentagon’s second designation. The ruling gives Anthropic relief, but the boundaries for government AI contracting are still being drawn.

Citation desk / original work

Cite this

Permanent attributionView citation
Finding 01

The important split is between procurement discretion and a government-wide security designation.

/posts/judge-vacates-pentagon-s-anthropic-blacklist-and-orders-defense-department-retraction#finding-1

Sources

  1. cnbc.comJudge blocks Pentagon blacklist of Anthropic as supply chain risk

Related coverage

Continue reading