Top AI Product

Anthropic just made it embarrassingly easy to leave ChatGPT behind. On March 1st, they quietly dropped [Claude Import Memory](https://claude.com/import-memory), a feature that lets you transfer all those personalized memories your current AI has built up about you — your name, your job, your coding preferences, the way you like your responses structured — straight into Claude. The whole thing takes about sixty seconds, and honestly, it feels like Anthropic just handed every ChatGPT power user an unlocked exit door.

Here’s how it works. You head over to [claude.com/import-memory](https://claude.com/import-memory), grab a pre-written prompt, and paste it into whatever AI you’re currently using. ChatGPT, Gemini, doesn’t matter. That prompt tells your old assistant to dump everything it knows about you into a neatly organized text block. You copy that output, paste it into Claude’s memory settings, and you’re done. No API keys, no file exports, no technical hoops. For ChatGPT users, there’s an even more direct route — just go to Settings, Personalization, Manage Memories, copy your entries, and paste them in.

The timing here is clearly intentional. This dropped the same week Claude shot to the top of the App Store following Anthropic’s very public clash with the Pentagon over AI ethics. The [Hacker News thread](https://news.ycombinator.com/item?id=47204571) titled “Switch to Claude without starting over” pulled in 371 points and 184 comments, with developers praising Claude’s code output quality and its refreshingly concise responses. Over on [Product Hunt](https://www.producthunt.com/products/claude-in-powerpoint), it hit #1 with 219 upvotes on the same day. [Tom’s Guide covered it](https://www.tomsguide.com/ai/claude-just-unlocked-memory-that-syncs-with-chatgpt-heres-how-it-works) too, calling it a major upgrade to how Claude handles personalization.

What makes this smart — and a little cheeky — is the privacy angle. Your imported memories are encrypted and explicitly not used for model training. Compare that to Google, which is reportedly testing a similar “Import AI Chats” feature for Gemini that imports full conversation histories and *will* use them for training. Anthropic is basically saying: “We’ll take your preferences, not your data.” That distinction matters, especially to the crowd already nervous about how these companies handle personal information.

There are limitations worth noting. Imported memories can take up to 24 hours to fully integrate since Claude processes them in daily synthesis cycles. And it only transfers memory-level context — not your conversation history, uploaded files, or any provider-specific features. But for most people, the memories are what actually matter. Nobody needs their old chat logs; they need their new AI to already know they prefer Python over JavaScript and hate bullet points in their summaries.

With ChatGPT sitting at 300 million weekly active users, even converting a small percentage is a massive win for Anthropic. And by making the switching cost basically zero, they’ve removed the single biggest reason people stay with an AI assistant they’ve outgrown. The feature is available on all paid plans — Pro at $20/month, Max, Team, and Enterprise. If you’ve been Claude-curious but dreading the cold start, that excuse just evaporated.