--- title: OpenAI Forced to Rewrite Pentagon Deal as 2.5 Million Users Join ChatGPT Boycott description: ChatGPT uninstalls surged 295% after OpenAI signed a Pentagon deal hours after Anthropic was blacklisted. Altman admitted it was 'opportunistic and sloppy'. The boycott, industry pushback and a court challenge followed. author: Darie Nani (Editor-in-Chief) date: 2026-03-06T11:30:00.000Z updated: 2026-03-27T21:04:01.344Z canonical: https://www.sovereignmagazine.com/article/openai-forced-to-rewrite-pentagon-deal-as-2-5-million-users-join-chatgpt-boycott image: https://cdn.nanimediahouse.com/1024px-Disrupt_SF_TechCrunch_Disrupt_San_Francisco_2019_-_Day_2_48838377432.jpg categories: Artificial Intelligence content_type: Analysis region: United States publication: Sovereign Magazine about: - type: Organization name: Anthropic - type: Organization name: OpenAI --- Anthropic's Claude reached number one on Apple's US App Store on 1 March, overtaking ChatGPT for the first time. It had been sitting at 42nd at the start of the year and had dropped outside the top 100 by late January. By the final week of February, daily signups were breaking all-time records. Free users increased by more than 60 per cent. Paid subscribers more than doubled. On 28 February alone, downloads jumped 51 per cent. The trigger was not a product launch or a price cut. Hours after President Trump [ordered federal agencies to stop using Anthropic](/science-tech/artificial-intelligence/anthropic-blacklisted-by-trump-after-rejecting-pentagon-ai-demands/) and Defence Secretary Pete Hegseth designated the company a 'supply chain risk to national security', Sam Altman announced that OpenAI had signed a deal to deploy its models on the Pentagon's classified network. Anthropic had been blacklisted for holding [two red lines](https://www.sovereignmagazine.com/article/judge-blocks-pentagon-anthropic-ban): no fully autonomous weapons and no mass surveillance of American citizens. OpenAI's published terms listed the same guardrails. But community notes on Altman's announcement, citing government officials, stated that the actual contract would allow the Pentagon to use OpenAI's models for 'all lawful purposes' — the same standard Anthropic had refused. ## 295 per cent uninstall spike US uninstalls of ChatGPT's mobile app jumped 295 per cent day-over-day on 28 February, according to Sensor Tower data, against a typical daily fluctuation of 9 per cent. Downloads dropped 13 per cent. One-star reviews surged 775 per cent in a single day, then doubled again on the Sunday. 'Cancel ChatGPT' trended across X and Reddit within hours. A Reddit post titled 'Cancel and Delete ChatGPT!!!' pulled more than 30,000 upvotes. A site called quitgpt.org launched and claimed more than 2.5 million people had taken action — cancelling subscriptions, pledging to stop using the app, or sharing the boycott. Those numbers include social media shares and website signups, not just verified cancellations, but protesters also gathered outside OpenAI's San Francisco headquarters. Anthropic launched a memory import tool in the same week, letting ChatGPT users transfer their saved conversations and preferences directly into Claude. ## OpenAI amends the contract On 3 March, Altman shared what he described as an internal memo on X. OpenAI would amend the Pentagon contract to include explicit language barring domestic surveillance of US persons and nationals, and prohibiting use by the NSA. He called his own deal 'opportunistic and sloppy' and said he 'shouldn't have rushed' to get it out on the Friday. He went further: 'To say it very clearly: I think this is a very bad decision from the DoW and I hope they reverse it.' Critics pointed out that the amended language was still vague, with carve-outs that could permit surveillance by intelligence agencies other than the NSA. OpenAI employees separately voiced support for Anthropic's position, according to CNBC. ## Four agencies drop Anthropic While OpenAI was rewriting its contract, the government expanded the blacklist. The departments of State, Treasury and Health and Human Services all dropped Anthropic's products, joining the Pentagon under the White House directive. Defence contractors were forced to certify they did not use Anthropic's technology. CNBC reported that defence tech companies were ditching Claude to comply, even where no viable alternative existed. Trump's own former AI advisor called the Pentagon's fight against Anthropic 'corporate murder'. Axios reported that the administration was treating DeepSeek, a Chinese AI company, more favourably than Anthropic, an American one built by former OpenAI researchers. ## Tech industry pushes back On 4 March, the Information Technology Industry Council — representing Apple, Adobe, Amazon, CoreWeave, Google, Meta, Microsoft and others — wrote to the White House urging Trump to reverse the designation. A second letter from the Software and Information Industry Association, TechNet, the Computer and Communications Industry Association and the Business Software Alliance raised the same concerns. Anthropic's technology is embedded in workflows across the tech ecosystem, including at companies that hold government contracts. Tech workers sent their own letter to the Department of Defence and Congress calling for the label to be withdrawn. ## Anthropic goes to court On 5 March, the Department of Defence officially notified Anthropic that the supply chain risk designation was in effect. Dario Amodei responded within hours: the company had 'no choice' but to challenge the designation in court. He argued the law requires the Secretary of Defence to use the least restrictive means necessary, and that a blanket designation fails that test. 'The threats do not change our position,' Amodei told CBS News the previous week. 'We cannot in good conscience accede to their request. We are patriotic Americans. Disagreeing with the government is the most American thing in the world.' Individual and commercial customers are unaffected by the designation. Anthropic's consumer app remains at number one on the App Store. ## Further Context **Q: Why are people boycotting ChatGPT?** OpenAI signed a deal to deploy its AI models on the Pentagon's classified network on 28 February 2026, hours after Anthropic was blacklisted by the Trump administration for refusing to remove safety restrictions on military use of its technology. Users saw OpenAI as profiting from a competitor's punishment. US uninstalls of the ChatGPT app jumped 295 per cent in a single day, and a boycott movement called QuitGPT claims more than 2.5 million participants. **Q: What is the ChatGPT Department of Defence deal?** OpenAI agreed to deploy its models within the Pentagon's classified network. The company published terms that included guardrails against autonomous weapons and mass surveillance, but community notes on Sam Altman's announcement cited government officials saying the actual contract allows use for 'all lawful purposes'. Altman later called the deal 'opportunistic and sloppy' and amended it to explicitly bar domestic surveillance and NSA use, though critics say the revised language still contains carve-outs. **Q: Can I cancel my OpenAI subscription?** Yes. Open the ChatGPT app or go to chat.openai.com, navigate to Settings, then Subscription, and select Cancel Plan. You can also request a full data export before deleting your account. Anthropic launched a memory import tool that lets former ChatGPT users transfer their saved conversations and preferences into Claude.