By Amrith Ramkumar, Heather Somerville and Chiqui Esteban

Anthropic overcame its latest spat with the Trump administration Tuesday, but new court documents released in one of its lawsuits against the Pentagon shed light on the underlying tensions between the government and one of America's hottest AI companies.

The documents show in detail the uneasy back-and-forth between Anthropic Chief Executive Officer Dario Amodei and Emil Michael, the undersecretary of defense for research and engineering. The pair are at the center of a landmark fight over appropriate guardrails for the Pentagon's use of artificial intelligence.

The new emails, disclosed in a court filing late Tuesday as part of the continuing litigation between the Defense Department and Anthropic, document how the relationship initially soured. They offer a deeper look into the flurry of communication between the company and the government before Anthropic was labeled a supply-chain risk and Pentagon partners were effectively barred from using its technology in their work with the Defense Department.

Below is a look at what the emails show.

Jan. 14, from Michael to Amodei

The dispute centered on Anthropic's opposition to the use of its models in domestic surveillance and AI-powered weapons, given the rapid pace of advancement in its models. The Pentagon insisted that it must have flexibility to make its own national-security decisions and should be able to use Anthropic's model Claude as it saw fit.

Michael reached out to Amodei in January to check in after the holidays following weeks of delay in the negotiations, which had been ongoing since late last year. Amodei's illness in December helped set the talks back.

Jan. 15, from Amodei to Michael

Amodei attempted to clarify the company's preference for bans on fully autonomous weapons and domestic surveillance. He focused on questions about how AI could aid the collection of data on Americans. He referred to the Department of War, using the unofficial name the Pentagon has revived for the Defense Department.

Feb. 4, from Michael to Amodei

Michael rejected Amodei's proposals and said the department insisted on using AI in all lawful-use cases. Michael said he didn't want to "force anything unnatural" if the two sides were too far apart on agreement; Amodei later offered the same sentiment.

Feb. 12, from Michael to Amodei

Amodei made another proposal to compromise on the weapons policy, but Michael pushed back. He cited an in-person discussion with Tarun Chhabra, Anthropic's head of national-security policy and a former Biden administration official who led work on technology and security.

Anthropic has disputed the Pentagon's characterization of some of the meetings and discussions.

Feb. 15, from Amodei to Michael

Amodei said following existing laws was inadequate because there are some legal ways to use AI for surveillance. He cited a section of legislation that allows surveillance for foreign-intelligence purposes. Sometimes those activities include communications of Americans.

Michael defended the Pentagon's practices, saying that they followed the law and that Anthropic in some cases was asking for language and compliance processes that went beyond what the Defense Department could implement.

Feb. 26, from Amodei to Michael

Amodei rejected the Defense Department's latest proposal, which he said added loopholes, and conceded there isn't a path forward to working together.

The following day, Defense Secretary Pete Hegseth said on social media that he was directing his department to label the company a supply-chain risk, which meant businesses working with the Pentagon could no longer use Anthropic's AI models. Anthropic has sued over the designation.

Scramble by other officials

The same day Hegseth targeted Anthropic, President Trump directed agencies to stop using the company's tools. A federal judge ordered the Trump administration to desist from applying the president's directive and from implementing aspects of the supply-chain risk designation. The government is appealing this decision while a lawsuit over other aspects of the designation plays out.

The documents show the race by federal agencies and offices to remove Claude quickly from their systems and find replacements. On March 3, shortly after 10 p.m., a Commerce Department official, Olivia Bradley, emailed others at the agency.

Michael recently said that two-thirds of the operations in the Pentagon that were using Anthropic's models have switched to other AI tools.

"While this case was necessary to protect Anthropic, our customers, and our partners, we remain focused on working productively with the government to ensure all Americans benefit from safe, reliable AI," an Anthropic spokeswoman said.

Write to Amrith Ramkumar at amrith.ramkumar@wsj.com, Heather Somerville at heather.somerville@wsj.com and Chiqui Esteban at chiqui.esteban@wsj.com