BEIJING — Chinese electric car company Nio is trying to raise the bar for premium vehicles in a fiercely competitive market.

Shares of Chinese electric carmaker Nio jumped as much as 10.45% in Hong Kong trading on Thursday after the company officially launched its ES9 SUV a day earlier. Its U.S.-listed stock closed 9.32% higher overnight, extending gains for 2026.

The ES9 starts at 390,000 yuan ($57,470) under Nio’s battery subscription model, which separates the cost of the vehicle from monthly battery payments.

It reflects the ongoing race to the bottom in China’s electric car market, despite Beijing’s efforts to curb excessive competition, often called involution. Sales of new energy vehicles for the first four months of the year have dropped by 17% in China, according to the country’s passenger car association.

The Chinese car market has already passed its years of fastest growth as most potential car buyers have already purchased one, Nio CEO William Li told reporters Thursday.

As technological features converge across EV models, Li said brand and targeted premiumization will be key to survival.

When Nio launched its flagship ET9 sedan in late 2023, prices started at 800,000 yuan. But before deliveries started in the first quarter of 2025, consumer electronics company Xiaomi had launched its first electric car — at 215,900 yuan.

With the new ES9, which Nio claims is the largest SUV in China, deliveries start Thursday.

CEO Li showed off an array of features at a launch event in Beijing, from advanced driver-assist systems that can respond to road signs to passenger seats with wood-colored tables that unfold similarly to those on an airplane. The ES9 also supports an in-car water boiler that lets passengers brew tea.

Nio signed on several brand promoters, including Robin Zeng, the CEO of CATL, the industry’s battery giant, who affirmed in a marketing video that about 2,000 of his employees had bought Nio cars.

Li also emphasized how the ES9 proactively protects passengers with “smart safety” systems that can detect and minimize impact from dangerous scenarios, and got China’s state broadcaster CCTV to livestream a crash test and other safety features.

Nio delivered 83,465 cars in the first quarter, nearly twice as many as a year ago, but a 33% drop from the fourth quarter. The figure also includes vehicles from Nio’s lower-priced brands Onvo and Firefly, which the company launched in the last two years to remain competitive in China’s sluggish consumer market.

Tesla‘s Model Y was the top-selling SUV in China last month by deliveries, according to industry data site China AutoHome. Elon Musk’s automaker last week received Beijing’s approval to launch driver assist in the country after years of waiting.

Nio’s ES8 ranked 10th in April deliveries across both electric and traditional gasoline-powered cars.

Foreign automakers are also revamping competition in China’s premium market at lower prices.

Audi on May 8 started presales for its E7X electric SUV with prices starting at 289,800 yuan, and is set to officially launch the car Friday morning. The car is the second model under the German automaker’s new China-focused brand, co-developed with Shanghai’s SAIC, that replaces the four-rings logo with the AUDI letters.

Like BYD and other Chinese automakers, Nio has pursued overseas expansion. But Li said Thursday that the automaker decided last year to shift its focus back to China.

He cited geopolitical tensions, including EU tariffs and the Russia-Ukraine war, as well as the far higher cost of investing in battery services overseas relative to China. Nio has largely focused much of its overseas expansion on Germany and Norway.

Underscoring Nio’s renewed domestic focus, Li said that the far-western Chinese region of Xinjiang represents a market twice the size of Norway’s

BEIJING — Electric carmakers in China are layering on more of the same artificial intelligence features as they try to survive a prolonged price war in the world’s largest auto market.

The competition has shifted over the last few years, from extending battery range to rolling out driver-assist systems and using more powerful automotive chips. Now, automakers are focusing on a suite of in-car AI features.

More than 50 car brands now use ByteDance’s Doubao AI model, the company’s cloud platform Volcano Engine announced last Friday at the Beijing auto show, where the tech unit had a booth next to robotaxi company Pony.ai.

That means Doubao is in 145 car models and over 7 million vehicles, Volcano Engine said. Besides domestic vehicles, Doubao AI has also been integrated in new foreign-branded models, such as the all-electric Mercedes-Benz GLC, the SAIC Audi E7X and the SAIC Volkswagen ID. ERA 9X.

“We will keep on integrating new features faster,” Fermín Soneira, CEO of the Audi and SAIC Cooperation Project, told reporters this month ahead of the auto show. He noted how automakers can quickly deploy tech updates remotely, or “over-the-air.”

Despite the rapid rollout of new features, automakers face persistent pressure on sales.

“It’s going to remain tough, because the capacity is there,” he said. “This price war is not going to really stop in the next month.”

The shift towards AI reflects consumer demand for connected features, including Huawei-smartphone-compatible interfaces or voice-based assistants such as Doubao.

ByteDance’s Doubao is by far the most widely used AI chatbot in China, with more than 155 million weekly active users as of early this year, according to consultancy Chozan. Volcano Engine’s auto show booth included demos of both Chinese-language and English-language AI systems for cars.

The price war has turned into a feature war around cockpit technology, said Stephen Dyer, partner and managing director and head of AlixPartners’ Asia automotive and industrials consulting practice.

The challenge is, however, that much of that technology soon becomes similar, making it harder for companies to stand out.

Among the top 20 best-selling electric car models in China, those priced at 100,000 yuan ($14,645) or above offered similar driver-assist and in-car entertainment functions, according to AlixPartners.

With “technology, they’re going to have to race and keep racing, because it disseminates so quickly that you’re never going to be able to sustain a differentiated technology for long,” Dyer said.

Instead, he expects Chinese companies to start competing more on the “outside-of-the-car experience,” similar to luxury brands that offer exclusive lifestyle experiences.

Chinese automaker Nio, for example, offers its customers exclusive access to products and clubhouses, on top of vehicles featuring premium interior materials.

The Chinese electric car company has struggled with the cost of offering such perks and slower market growth. But Nio claimed last week its ES8 is the first car model in the industry’s 400,000 yuan-and-above segment to deliver 100,000 units in just 215 days.

Alibaba also announced Friday that its Qwen artificial intelligence model will be integrated into vehicles from automakers including BYD and a local joint venture of Volkswagen. The system allows drivers to order food delivery, book hotels, buy tickets to attractions and track packages, among other features, through voice commands.

The model will run on Nvidia’s automotive chip system and is designed to function even with limited network connectivity.

At the end of the day, AI should run in the background to support the user experience, not necessarily be a feature of a vehicle, Tu Le, founder and managing director at consultancy Sino Auto Insights, told CNBC’s Eunice Yoon.

Even if it’s difficult for automakers to stand out in China, they may be able to compete more effectively with foreign peers.

“What we consider maybe simple features and like, standard features in mass market vehicles in the China market, are going to be expected in the Western market sooner rather than later as well,” Le said.

A federal appeals court in Washington, D.C., on Wednesday denied Anthropic’s request to temporarily block the Department of Defense’s blacklisting of the artificial intelligence company as a lawsuit challenging that sanction plays out.

The ruling comes after a judge in San Francisco federal court late last month, in a separate but related case, granted Anthropic a preliminary injunction that bars the Trump administration from enforcing a ban on the use of its Claude model.

“In our view, the equitable balance here cuts in favor of the government,” the appeals court said in its decision. “On one side is a relatively contained risk of financial harm to a single private company. On the other side is judicial management of how, and through whom, the Department of War secures vital AI technology during an active military conflict. For that reason, we deny Anthropic’s motion for a stay pending review on the merits.”

With the split decisions by the two courts, Anthropic is excluded from DOD contracts but is able to continue working with other government agencies while litigation plays out. Defense contractors will be prohibited from using Claude in their work with the agency, but they can use it for other cases.

The DOD declared Anthropic a supply chain risk in early March, meaning that use of the company’s technology purportedly threatens U.S. national security. The label requires defense contractors to certify that they don’t use Anthropic’s Claude AI models in their work with the military.

Anthropic had asked the appeals court to review the Pentagon’s determination and argued that it’s a form of retaliation that’s unconstitutional, arbitrary, capricious and not in accord with procedures required by law, according to a filing. 

In the ruling on Wednesday, the court acknowledged that Anthropic “will likely suffer some degree of irreparable harm absent a stay,” but that the company’s interests “seem primarily financial in nature.” While the company claimed the DOD was standing in the way of its right to free speech, “Anthropic does not show that its speech has been chilled during the pendency of this litigation,” the order said.

Because of the harm Anthropic is likely to suffer, the appeals court said “substantial expedition is warranted.”

An Anthropic spokesperson said in a statement after the ruling that the company is “grateful the court recognized these issues need to be resolved quickly” and that it’s “confident the courts will ultimately agree that these supply chain designations were unlawful.”

“While this case was necessary to protect Anthropic, our customers, and our partners, our focus remains on working productively with the government to ensure all Americans benefit from safe, reliable AI,” Anthropic said.

Todd Blanche, the acting U.S. attorney general, called the decision a “resounding victory for military readiness,” in a post on X.

“Military authority and operational control belong to the Commander-in-Chief and Department of War, not a tech company,” Blanche wrote.

The DOD relied on two distinct designations under the U.S. federal court to justify the supply chain risk action, and they have to be challenged in two separate courts.

Anthropic’s suit against the Pentagon in March followed a dramatic couple weeks in Washington D.C., between the Department of Defense and one of the most valuable private companies in the world.

In a post on X in late February, Defense Secretary Pete Hegseth declared Anthropic a supply chain risk, and the DOD soon notified the company of the official determination via a letter. Anthropic is the first American company to be given the designation, which has historically been reserved for foreign adversaries.

Shortly before Hegseth’s post, President Donald Trump wrote a Truth Social post ordering federal agencies to “immediately cease” all use of Anthropic’s technology. He said there would be a six-month phase-out period for agencies like the DOD.

The Trump administration’s actions surprised many officials in Washington, where Anthropic’s technology had become embedded in numerous agencies. The company was the first to deploy its models across the DOD’s classified networks, and it was championed for its ability to integrate with existing Defense contractors like Palantir. 

Anthropic signed a $200 million contract with the Pentagon in July, but as the company began negotiating Claude’s deployment on the DOD’s GenAI.mil AI platform in September, talks stalled.

The DOD wanted Anthropic to grant the Pentagon unfettered access to its models across all lawful purposes, while Anthropic wanted assurance that its technology would not be used for fully autonomous weapons or domestic mass surveillance. 

The two failed to reach an agreement, pushing the dispute to court.

— CNBC’s Dan Mangan contributed to this report.

WATCH: Anthropic wins preliminary injunction in fight over Pentagon blacklisting

Palantir CEO Alex Karp told CNBC on Thursday that artificial intelligence is giving the U.S. and its allies an edge in the escalating conflict in Iran and across the Middle East.

“What makes America special right now is our lethal capabilities, our ability to fight war,” Karp said at Palantir’s AIPcon 9 in Maryland. He added that another major advantage is that “the AI revolution is uniquely American.”

Karp alluded to his company’s ability to link combat data between the U.S. and Middle East partners that were hit by Iranian airstrikes.

“If you were attacked and you needed to coordinate, you would have to have a coordinating function. There’s only one product that can actually do that for security,” Karp said, referencing Palantir’s platform.

Palantir’s Project Maven is a real-time AI surveillance capability that leverages satellite imagery. The platform was used with Anthropic’s Claude in the capture of Venezuela’s President Nicolás Maduro, according to the Wall Street Journal.

Karp declined to comment on whether Maven was used to kill Iranian Supreme Leader Ayatollah Ali Khamenei in a joint U.S.-Israel military operation two weeks ago.

“I have read that Palantir’s Project Maven is the core backbone of that,” he said, speaking generally about U.S. involvement in the Middle East. “And then I’ve also read that all the allies, Arab and non-Arab in the Middle East, may or may not be users of our platform as well, and that’s expanding rapidly.”

Experts and executives in the industry say AI is pushing the conflict to a new frontier. That was evident when Iran bombed three Amazon data centers in the Middle East last week.

U.S. data centers are increasingly seen as national security assets, hosting critical digital infrastructure used by governments and big companies.

“They’re evil, they’re not stupid,” Karp said. “Look who’s on the list, look who’s not. We’re in the middle of war. You would expect it to be a list of hardcore military companies. They are interested in the things they can’t produce.”

While Palantir may be best known for its defense technology, the company’s commercial business is booming. U.S. commercial revenue jumped 137% in the fourth quarter to $507 million.

Palantir shares are up 12% so far this month, while the Nasdaq is down about 1.6%.

WATCH: Palantir’s Q4 earnings beat

OpenAI CEO Sam Altman met with a handful of lawmakers in Washington, D.C. where Sen. Mark Kelly, D-Ariz., said he raised some “serious questions” about the company’s approach to warfare and its recent deal with the Department of Defense.

In an interview with CNBC’s Emily Wilkins, Kelly said the group talked “in detail” about surveillance and how artificial intelligence systems could be used within a kill chain. He called it a “good discussion.”

“There’s got to be guardrails in place, and we’ve got to make sure that we’re always thinking about the Constitution and making sure that we comply with it,” Kelly said.

OpenAI formed a deal with the DOD late last month just hours after rival Anthropic had been blacklisted by Defense Secretary Pete Hegseth, who declared the company a “Supply-Chain Risk to National Security.”

Anthropic had been trying to renegotiate its contract with the DOD, but the talks stalled over a disagreement about how the technology could be used. The DOD wanted Anthropic to grant the military unfettered access to its models for all lawful purposes, while Anthropic sought assurance that its models would not be used for fully autonomous weapons or domestic mass surveillance.

Altman said in a post on X the day the deal fell apart that prohibitions on domestic mass surveillance and human responsibility for the use of force, including for autonomous weapon systems, are two of the company’s “most important safety principles.” He said the DOD agreed and put them into the arrangement.

OpenAI published an excerpt of its contract with the DOD, which says that the agency “may use the AI System for all lawful purposes.” The company said it’s confident the DOD will not be able to use its AI systems for mass surveillance or fully autonomous weapons because of OpenAI’s safety stack, the contract language and existing laws.

“We think it’s very important to support the United States government and the democratic process,” Altman told CNBC on Thursday. He added that while OpenAI disagrees with the Pentagon’s decision to designate Anthropic a supply chain risk, he thinks the government needs to be able to make decisions about “how the most important things in the country are going to work.”

The clash between Anthropic and the DOD came as a shock to officials and technologists in Washington. Many had come to view Anthropic’s models as superior — they were the first to be deployed in the agency’s classified networks — and championed the company’s ability to integrate with existing defense contractors like Palantir. 

Kelly is working with other senators to draft legislation that would set guardrails around DOD contracts with AI organizations, and said that Congress “needs to have a role.”

“We need to have legislation on this that creates some of these boundaries, guardrails,” he said. “This is the United States Congress, things don’t move as fast as we would like, but this technology is moving very quickly.”

WATCH: Sam Altman on meeting with lawmakers: Economic impact on jobs will be a huge topic

Palantir is still using Anthropic’s Claude as the artificial intelligence startup’s clash with the Pentagon plays out, CEO Alex Karp told CNBC Thursday.

“The Department of War is planning to phase out Anthropic; currently, it’s not phased out,” Karp told CNBC’s Seema Mody at Palantir’s AIPcon 9 in Maryland. “Our products are integrated with Anthropic, and in the future, it will probably be integrated with other large language models.”

The Department of Defense officially designated Anthropic a supply-chain risk last week, but is still using Claude models to support the war in Iran, as CNBC previously reported.

Anthropic sued the Trump administration on Monday to reverse the supply chain risk designation and is seeking a stay on the DOD action.

This is the first time that Palantir has commented since the Pentagon’s designation. Anthropic and Palantir partnered with Amazon Web Services to support the DOD in 2024.

Other defense tech companies have told employees to stop using Claude as the clash plays out, including Lockheed Martin.

Defense Department CTO Emil Michael told CNBC Thursday that it would take time for the government to transition away from Anthropic’s models.

“You can’t just rip out a system that’s deeply embedded overnight,” Michael told CNBC’s “Squawk Box.”

Last August, Pentagon technology chief Emil Michael, a former Uber executive and attorney, took on the added role of overseeing the Defense Department’s artificial intelligence portfolio. A month earlier, Anthropic had been awarded a $200 million DOD contract that expanded its work with the agency.

“I said, ‘I just want to see the contracts,'” Michael told the All-In Podcast on Friday, reflecting on his early days managing the AI portfolio. “You know, the old lawyer in me.”

Michael’s request kicked off a months-long review process that culminated in the Defense Department officially banning Anthropic’s technology last week, leaving the military without its hand-picked AI models to operate in the most sensitive environments. Anthropic was booted for demanding that its AI not be used for autonomous weapons or domestic surveillance.

In an extraordinary move, the DOD designated Anthropic a supply chain risk, a label that’s historically only been applied to foreign adversaries. It will require defense vendors and contractors to certify that they don’t use the company’s models in their work with the Pentagon.

Anthropic sued the Trump administration on Monday, calling the government’s actions “unprecedented and unlawful,” and claiming that they are “harming Anthropic irreparably,” putting hundreds of millions of dollars worth of contracts in jeopardy.

The DOD’s sudden reversal came as a shock to many officials in Washington who viewed Anthropic’s models as superior — they were the first to be deployed in the agency’s classified networks — and championed the company’s ability to integrate with existing defense contractors like Palantir. The decision was all the more puzzling since the Trump administration had threatened during negotiations to invoke the Defense Production Act, which could have forced Anthropic to grant the military access to its technology.

“I don’t know how those two things can both be true in reality,” said Mark Dalton, a retired Navy rear admiral who now leads technology and cybersecurity policy at R Street, a think tank in Washington, D.C. “Something is so necessary that you need to invoke DPA and so harmful that you put a designation on it that’s reserved for foreign adversaries.”

Defense experts like Dalton expressed concern about the government’s decision. Not only does it set a troubling precedent, they argue, but it also means the administration is banishing a key technology vendor that’s been lauded for its diligence with respect to AI safety, tough rhetoric against China and its entrepreneurial chops, becoming one of the fastest-growing tech startups in the U.S. 

Former DOD official Brad Carson, who’s now co-founder and president of AI policy nonprofit Americans for Responsible Innovation, said the move is particularly troubling for military personnel, who have come to rely on Anthropic’s Claude models. An ex-Navy intelligence officer who served in Iraq, Carson said he’s talked to a number of retired officers who told him that “warfighters are not happy about it.”

“You’re not so excited if you’re in the military,” said Carson, who worked in President Obama’s Defense Department until 2016 and before that was deployed to Iraq while in the Army and also served two terms in Congress as a Democrat in Oklahoma. “They view Claude as being a better product, the most reliable, with the most user friendly outputs they can assimilate into planning.”

Anthropic declined to provide a comment for the story.

According to a senior department official at the DOD, vendors will not be allowed to insert themselves “into the chain of command by restricting the lawful use of a critical capability.”

“It is the military’s sole responsibility to ensure our warfighters have the tools they need to win in a crisis, without interference from corporate policies,” the official said.

CNBC spoke to 17 AI policy experts, former Palantir and Anthropic employees, tech analysts and researchers about Anthropic’s critical role in the Defense Department and what comes next. Several of the people asked not to be named because they weren’t authorized to speak on the matter.

Anthropic CEO Dario Amodei founded the San Francisco-based company in 2021 alongside his sister, Daniela Amodei, and a handful of other researchers. The group had defected from OpenAI, before the launch of ChatGPT, over concerns about the company’s direction and attitude toward safety. They spent years carefully constructing Anthropic’s reputation as a firm that was more dedicated to responsible AI deployment. 

Anthropic launched its family of AI models, known as Claude, in March 2023, a few months after ChatGPT hit the market and quickly went viral. In the three years since introducing Claude, Anthropic has raised billions of dollars of capital, en route to a $380 billion valuation. 

The company is now under immense pressure to justify that price tag and has been forced to rapidly commercialize its technology in an effort to keep pace with OpenAI and other rivals like Google.

Amazon has been one of Anthropic’s largest financial backers since 2023, investing a total of $8 billion in the startup.

As pilot projects got underway, many federal employees found that Claude produced more compelling results than other models from companies like OpenAI and Meta, the sources said. Claude could provide step-by-step reasons for why it would derive an answer or complete a task, which was crucial for federal agencies that require strong auditing and verification, sources said.

And since Anthropic had prioritized building for enterprise customers, the company’s user experience was especially suitable for desktop computers, the people said. With a strong AI model and an intuitive user interface, Anthropic began to earn credibility with federal workers, sowing the seeds for a major partnership with Palantir, a software and services provider that counts on government contracts for about 60% of its U.S. revenue.

Anthropic’s government push was assisted by its head of global affairs, Michael Sellitto, who led cybersecurity policy at the National Security from 2015 through 2018, and former AWS executive Thiyagu Ramasamy, the head of the company’s public sector business.

In a LinkedIn post last week, Ramasamy, who joined in early 2025, expressed his concern about the Pentagon’s actions.

“Today, I mourn for the customers I have deep respect for,” Ramasamy wrote. “They were moving at a pace I couldn’t have imagined across my two decades in this industry, and now it comes to a halt over a weekend. Down, but not out.”

And in a legal filing on Monday tied to Anthropic’s lawsuit, Ramasamy said he has a team of 15 people working with federal government customers, and the company’s public sector business was projected to reach multiple billions of dollars in annual recurring revenue within five years.

“The Government’s Actions are an existential threat to all of this,” Ramasamy said in the filing.

Sellitto and Ramasamy didn’t respond to requests for comment.

In November of 2024, shortly before Ramasamy’s arrival, Anthropic and Palantir announced a partnership with AWS that would allow U.S. intelligence and defense agencies to access Claude. Some Anthropic staffers were upset about the deal when it was announced, a former employee said, adding that it prompted “many big Slack threads” and became a point of lingering tension within the company. 

Having Palantir as a partner helped Anthropic build direct lines with the DOD and fast tracked its integration into the highest-level, classified projects. The partnerships were crucial in helping Anthropic become the first model company to officially deploy across classified networks, said Lauren Kahn, a senior research analyst at Georgetown’s Center for Security and Emerging Technology.

“The fact that Anthropic is able to basically play nice with others like Palantir, AWS, Google, etc., specifically Palantir,” she said, “is extremely valuable.”

In its July 2025 release announcing its $200 million defense contract, Anthropic said it had “accelerated mission impact across U.S. defense workflows with partners like Palantir.” Anthropic said its technology helped the government’s defense and intelligence organizations “rapidly process and analyze vast amounts of complex data.”

A month after winning the Pentagon contract, Anthropic partnered with the U.S. General Services Administration to bring its AI models to other participating agencies for $1 dollar a year.

Amodei told staffers earlier this month that the administration doesn’t like Anthropic because it hasn’t donated or offered “dictator-style praise to Trump,” according to a report from The Information. 

He apologized for the tone of those remarks in a statement on Thursday, writing that they were written after a “difficult day for the company” and do not “reflect my careful or considered views.”

Amodei has also drawn the ire of David Sacks, the the White House AI and crypto czar, who has accused Anthropic of supporting “woke AI,” largely for its positions on regulation.

“This feels to me like a dispute that is about politics and personalities,” Michael Horowitz, a senior fellow for technology and innovation at the Council on Foreign Relations, said in an interview. “It’s masquerading as a policy dispute.”

A DOD official told CNBC that the decision is not based on personality or politics but rather about “the military being able to use technology for all lawful purposes.”

By the time the Trump administration blacklisted Anthropic, the startup’s tools had been widely adopted across government agencies. A transition is already underway, as groups including the U.S. Department of Health and Human Services, the Treasury Department and the State Department have confirmed they are moving off of Claude.

But that process is especially complicated within the DOD, in part because the U.S. is actively carrying out a military operation in Iran. Anthropic’s models have been used to support that operation, even after it was blacklisted, as CNBC previously reported. 

Amodei said in a statement on Thursday that Anthropic’s “most important priority” is making sure American warfighters and national security experts aren’t deprived of tools in the middle of “major combat operations.” The company has committed to offering its models at a nominal cost with continuing support for engineers “for as long as is necessary to make that transition, and for as long as we are permitted to do so.”

Moving away from Anthropic toward a new vendor will take the DOD time and comes at a significant cost in terms of efficiency, said Jacquelyn Schneider, a Hargrove Hoover fellow at Stanford University’s Hoover Institution, in an interview.

“You’re not going to walk away from technologies that are deeply embedded in your wartime processes right before you go to war,” Schneider said. 

WATCH: Why the U.S. Defense Department blacklist of Anthropic is so unprecedented

Anthropic filed a lawsuit against the Trump administration on Monday after the artificial intelligence startup was blacklisted and labeled a threat to U.S. national security.

The company said in a complaint that these actions are “unprecedented and unlawful,” and that they are “harming Anthropic irreparably.” The lawsuit was filed in the U.S. District Court for the Northern District of California.

“Anthropic’s contracts with the federal government are already being canceled. Current and future contracts with private parties are also in doubt, jeopardizing hundreds of millions of dollars in the near-term,” the filing says. “On top of those immediate economic harms, Anthropic’s reputation and core First Amendment freedoms are under attack. Absent judicial relief, those harms will only compound in the weeks and months ahead.”

The lawsuit is the latest episode in a dramatic two-week saga between Anthropic and the Trump administration over how the company’s AI models can be used on the battlefield and elsewhere. Before the spat between the two sides escalated into the public’s view late last month, Anthropic served as an early partner across many U.S. agencies as the government sought to rapidly upgrade its systems and capabilities with cutting-edge AI technology.

On Thursday, Anthropic confirmed that it had officially been designated a supply chain risk, an extraordinary move that has historically been reserved for foreign adversaries. It will require defense vendors and contractors to certify that they don’t use Anthropic’s models in their work with the Pentagon.

President Donald Trump also shared a post on social media last month directing federal agencies to “immediately cease” all use of Anthropic’s technology.

“WE will decide the fate of our Country — NOT some out-of-control, Radical Left AI company run by people who have no idea what the real World is all about,” Trump wrote.

Anthropic has asked the court to vacate the supply chain risk designation and to grant the company a stay on the action as the legal case unfolds.

Anthropic signed a $200 million contract with the Department of Defense in July, and it was the first AI lab to deploy its technology across the agency’s classified networks. The company had been renegotiating the going forward terms of its contract, and talks hit a snag because the two organizations disagreed over how its models could be used.

The DOD wanted Anthropic to grant the agency unfettered access to its AI models across all lawful purposes, while Anthropic wanted assurance that its models would not be used for fully autonomous weapons or domestic mass surveillance.

Anthropic’s models have still been used to support the U.S.’ military operation in Iran, even after the company was blacklisted, as CNBC previously reported.

“Seeking judicial review does not change our longstanding commitment to harnessing AI to protect our national security, but this is a necessary step to protect our business, our customers, and our partners,” an Anthropic spokesperson told CNBC on Monday. “We will continue to pursue every path toward resolution, including dialogue with the government.”

A spokesperson for the DOD said the agency does not comment on litigation.

More than a dozen federal agencies, including the DOD, the U.S. Department of the Treasury, the U.S. Department of State and the General Services Administration, are named as defendants in the lawsuit.

“The consequences of this case are enormous,” the complaint says. “Defendants are seeking to destroy the economic value created by one of the world’s fastest-growing private companies, which is a leader in responsibly developing an emergent technology of vital significance to our Nation.”

WATCH: Why the U.S. Defense Department blacklist of Anthropic is so unprecedented

Anthropic CEO Dario Amodei confirmed that the U.S. government declared his company a supply chain risk on Thursday and said it has “no choice” but to challenge the designation in court.

The startup has been at odds with the Department of Defense over how its artificial intelligence models, known as Claude, can be used, and was told late last week, via social media posts, that it was being blacklisted from government contracts.

Anthropic sought assurance that its technology would not be tapped for fully autonomous weapons or domestic mass surveillance, but the DOD wanted Anthropic to grant the agency unfettered access to Claude across all lawful purposes.

“As we stated last Friday, we do not believe, and have never believed, that it is the role of Anthropic or any private company to be involved in operational decision-making—that is the role of the military,” Amodei wrote. “Our only concerns have been our exceptions on fully autonomous weapons and mass domestic surveillance, which relate to high-level usage areas, and not operational decision-making.”

Anthropic is the only American company ever to be publicly named a supply chain risk, and the designation, which is now official, will require defense vendors and contractors to certify that they don’t use the company’s models in their work with the Pentagon. The label has typically been reserved for organizations that operate within foreign adversaries, like Chinese tech company Huawei. 

Uncertainty remains as to whether defense contractors can use Anthropic’s technology for projects outside of their work with the military. Amodei said in his post that the designation “doesn’t (and can’t) limit uses of Claude or business relationships with Anthropic if those are unrelated to their specific Department of War contracts.”

Microsoft, which announced plans to invest up to $5 billion in Anthropic in November, said in a statement that its lawyers “studied the designation” and determined that Anthropic products can remain available to its customers other than the DOD.

Anthropic signed a $200 million contract with the DOD in July, and it was the first AI lab to integrate its models into mission workflows on classified networks. But as negotiations between the two sides stalled, rivals OpenAI and Elon Musk’s xAI also agreed to deploy their models in classified capacities.

OpenAI CEO Sam Altman announced his company’s deal with the DOD hours after Anthropic was blacklisted on Friday. He said in a post on X that the agency displayed a “deep respect for safety and a desire to partner to achieve the best possible outcome.”

Anthropic’s relationship with the Trump administration has grown increasingly tense in recent months, and Amodei apologized for a critical internal memo that was leaked to the press on Wednesday.

Amodei reportedly told staffers that the administration doesn’t like Anthropic because it hasn’t donated or offered “dictator-style praise to Trump,” according to a report from The Information.

He said the memo was written on Friday after a “difficult day for the company” and does not reflect his “careful or considered views.” Amodei added that it is an “out-of-date assessment of the current situation.”

“Anthropic did not leak this post nor direct anyone else to do so—it is not in our interest to escalate this situation,” Amodei wrote.

WATCH: Sec. Pete Hegseth directs Pentagon to designate Anthropic supply-chain risk to national security

Amazon said Friday it will continue offering Anthropic’s artificial intelligence technology to its cloud customers, excluding work involving the Department of Defense.

The announcement comes after the federal agency informed Anthropic on Thursday that it would label the company a “supply chain risk.” Anthropic responded by saying it has “no choice” but to challenge the designation in court.

“AWS customers and partners can continue to use Claude for all their workloads not associated with the Department of War (DoW),” an Amazon Web Services spokesperson said in a statement. “For all DoW workloads which use Anthropic technologies, we are supporting customers and partners as they transition to alternatives running on AWS.”

Last week, President Donald Trump ordered federal agencies, via social media posts, to stop using Anthropic’s technology. The declaration came after Anthropic refused to agree to the DOD’s request that it be allowed to operate the company’s technology in all lawful use cases without limitation.

The supply chain risk designation will require defense vendors and contractors to certify they don’t use Anthropic’s models in their work with the Pentagon.

Amazon, the leader in public cloud, is following top rivals Microsoft and Google in updating customers on Anthropic’s availability. Microsoft said late Thursday that Anthropic’s Claude models will remain accessible in its products outside of defense work, and Google issued a similar statement earlier Friday.

Amazon is one of Anthropic’s biggest financial backers, investing $8 billion in the startup since 2023. The two companies have also forged a strong commercial relationship.

AWS remains Anthropic’s primary cloud and training partner. Anthropic also committed to use 500,000 of Amazon’s custom-built chips, called Trainium 2, as part of an $11 billion AWS data center campus built for the startup, called Project Rainier.

Anthropic’s Claude models are available through AWS Bedrock, which allows companies to tap into artificial intelligence models built by a variety of providers. AWS offers Bedrock to users of its GovCloud service, a dedicated cloud region equipped to host sensitive data and regulated workflows.

Amazon has won billions of dollars in contracts to provide cloud and AI services to federal agencies. The company in November earmarked up to $50 billion on AI infrastructure for U.S. government customers. At the time, AWS said it served more than 11,000 government agencies.

Anthropic relied on AWS as its pathway into government work. In November 2024, the company partnered with software and services vendor Palantir and AWS to provide defense and intelligence agencies access to Claude models.

Dave Levy, an AWS vice president, said in the announcement that, “We are excited to partner with Anthropic and Palantir and offer new generative AI capabilities that will drive innovation across the public sector.”

Eight months later, Anthropic was awarded a $200 million DOD contract and became the first AI lab to integrate its models into mission workflows on classified networks.

— CNBC’s Ashley Capoot contributed to this report.