Lutnick Says Anthropic Is Back on Trump Administration’s “Right Side”

Commerce Secretary Howard Lutnick said Anthropic had repaired its relationship with the Trump administration after months of conflict over how the Pentagon could use the company’s Claude AI models. In comments reported September 2, 2026, Lutnick said the administration trusted Anthropic and that the company was “back on the right side.” 13

What Lutnick said about Anthropic

Lutnick’s remarks signaled a political thaw rather than a newly announced policy or contract decision. In an Axios interview, he said, “We trust Anthropic,” adding that the company had “done what we asked” and was “back on the right side.” 13

Speaking separately at the G20 technology gathering in North Carolina, Lutnick characterized the earlier conflict as a public “kerfuffle” over security issues and said Anthropic had “gotten religion.” 2

Why Anthropic had clashed with the Pentagon

The dispute centered on Anthropic’s limits on use of Claude. Defense Secretary Pete Hegseth had blocked the company from certain military contracts after Anthropic would not allow its models to be used for domestic surveillance or autonomous weapons. 67

Anthropic responded by suing in California federal court over the Pentagon’s designation of the company as a supply-chain risk. 4

The court ruling on the Pentagon’s blacklisting

On August 27, U.S. District Judge Rita Lin blocked the Pentagon’s blacklisting action. Reuters reported that Lin found the designation “illegal and baseless.”

Other reporting described the ruling as finding that the Pentagon’s action was unlawful retaliation and that Anthropic had been denied required due process. The decision was a significant legal rebuke of the government’s approach to the dispute, even as Lutnick’s subsequent comments suggested that the broader relationship was improving.

Tom Brown’s G20 message: build data centers

Anthropic co-founder Tom Brown appeared at the G20 technology event in Chapel Hill, North Carolina, on the same day. His reported emphasis was on AI infrastructure: he urged officials to grant permits for data centers in their countries.

That focus matters because the episode combined two separate but connected AI policy questions: what constraints should apply when advanced models are used in sensitive military contexts, and whether governments will enable the computing infrastructure needed to develop and operate those systems at scale. Lutnick’s comments indicated warmer administration ties, while the court ruling showed that the legal limits on government pressure remained consequential.