OpenAI is staring at mounting criticism and potential subscription losses after entering into a partnership with the US government to provide artificial intelligence tools for Pentagon-linked projects. The agreement, announced last week, has triggered a wave of online backlash, with some users calling for a boycott of ChatGPT over the company’s involvement in military initiatives.

The controversy intensified after US Defense Secretary Peter Hegseth reportedly classified rival AI firm Anthropic as a “supply chain risk.” The designation came after Anthropic declined to grant the Pentagon unrestricted access to its AI systems for military applications. The timing of these developments — with OpenAI’s deal emerging just hours later — has fueled speculation and criticism online.

Social media platforms such as Reddit and X have since seen a surge of posts from users claiming they plan to switch from ChatGPT to Anthropic’s Claude chatbot. Adding momentum to the protest, a newly launched boycott website alleges that more than 2.5 million users have pledged to stop using ChatGPT in response to the Department of Defense agreement.

“ChatGPT takes Trump's killer robot deal. It's time to Quit.” reads the website’s headline message.

The platform says its mission is to mobilize “Americans and people around the world to quit ChatGPT.” It further argues that many ChatGPT users are unaware of alternative AI tools and encourages them to explore other options.

“ChatGPT users skew young and progressive, and many don't know about alternatives. We can push OpenAI over the edge. If we make an example of ChatGPT, we can send a clear signal to ICE enablers that their actions will not go unpunished. Let's make CEOs think twice before they get in bed with Trump.” reads the website.

In addition to targeting ChatGPT, the boycott call extends to Elon Musk’s Grok chatbot, while promoting open-source AI alternatives such as Confer, Alpine, and Lumo. It also suggests corporate competitors like Gemini and Claude as substitutes.

Data cited in a recent report by TechCrunch, referencing market intelligence firm Sensor Tower, indicates a notable spike in user churn. According to the report, ChatGPT mobile app uninstalls jumped 295% on Saturday — just one day after the Pentagon partnership was made public. Meanwhile, downloads of Claude in the US rose sharply, climbing 37% on Friday and surging 51% on Saturday.

Amid the criticism, OpenAI CEO Sam Altman has publicly defended the company’s decision. Speaking both in private discussions and on social media, Altman acknowledged that the rollout of the announcement may have been mishandled.

“The issues are super complex, and demand clear communication. We were genuinely trying to de-escalate things and avoid a much worse outcome, but I think it just looked opportunistic and sloppy. Good learning experience for me as we face higher-stakes decisions in the future.” he wrote on X.

As debates around AI ethics, national security, and corporate responsibility intensify, OpenAI now finds itself navigating one of the most significant public relations challenges in its history.