Why ChatGPT Uninstalls Jumped 295% and Claude’s One‑Click Migration Turned the Tide

Anthropic’s Claude reached number one on the US App Store free list, pushing the long‑dominant ChatGPT to second place. Data shows ChatGPT’s daily uninstall volume in the US surged 295% in a single day (usual daily fluctuation is about 9%). Claude’s free‑user base grew over 60% since January, its paid subscriptions doubled, daily new registrations hit record highs, and the QuitGPT protest movement claims more than 1.5 million participants.

The catalyst was a February end request from the US Department of Defense for Anthropic to allow its models to be used for any legal purpose, including large‑scale domestic surveillance and autonomous weapons. Anthropic’s CEO Dario Amodei refused on conscience grounds, leading the company to be placed on a “supply‑chain risk” list and federal agencies were instructed to stop using its products. On the same day, OpenAI’s CEO Sam Altman announced a $200 million contract with the Pentagon to deploy ChatGPT models on classified military networks, praising the department’s “deep respect for safety.”

The public reaction exploded: Reddit threads titled “Cancel ChatGPT” flooded the platform, screenshots of subscription cancellations spread on X (formerly Twitter), singer Katy Perry posted a Claude Pro subscription screenshot with the word “done,” over 700 Google and OpenAI employees signed an open letter supporting Anthropic, and the QuitGPT group planned a protest at OpenAI’s San Francisco headquarters on March 3.

Compounding the backlash, OpenAI introduced ads into ChatGPT on February 9, showing sponsored content to free and low‑tier users, sometimes even in the first reply. Although OpenAI claimed the ads fund free service operations and do not affect paid users, many users complained, “I’m chatting with AI and it’s trying to sell me stuff.” In contrast, Anthropic announced an “never‑place‑ads” policy early February, even buying a Super Bowl ad spot to mock AI‑in‑chat advertising, stating its revenue comes solely from enterprise contracts and paid subscriptions reinvested into Claude.

Amid the influx of users, Anthropic executed two precise moves. First, it opened the Memory feature—previously exclusive to paid users—to all free users on March 2, allowing Claude to retain user preferences, work habits, and technical stacks across sessions permanently. Second, it launched a “memory import” tool that lets users migrate all their context from ChatGPT (or any other AI) to Claude with a single prompt.

The migration process consists of three steps:

Visit claude.com/import-memory and copy the provided prompt, which asks the source AI to list every stored memory and context in a single code block.

Paste the prompt into the ChatGPT (or other AI) conversation and send it, prompting the AI to output all personal data, preferences, projects, tools, and behavior corrections.

Copy the resulting code block and paste it into Claude’s Settings → Capabilities → Memory, then click “Add to memory.”

Anthropic emphasizes that “Context shouldn’t disappear because you want to try something new,” and Claude also supports exporting your memory data at any time, reinforcing a product philosophy that puts user data control at the forefront.