If you've been using ChatGPT for a while, it knows a lot about you. Your writing style, your work, how you like things explained. Switching tools feels like giving all of that up and starting over — which is honestly why most people don't bother.
I almost didn't either.
But Claude has a memory import feature that lets you pull everything ChatGPT knows about you and bring it straight into Claude. The whole thing takes under 5 minutes. I timed it.
Here's how it works.
First — Why Claude at all?
Fair question. ChatGPT is good. I still use it.
But after spending time with Claude, a few things genuinely surprised me. It handles long documents without losing the thread halfway through. It's more careful about admitting when it doesn't know something, which matters a lot if you're teaching people and don't want to pass on nonsense. And for writing and explaining things in plain language — which is basically my whole job — it just clicks.
You don't have to pick one. I use both, depending on what I need. Different tools for different jobs.
What You Need
That's it.
The Steps
Step 1. Open claude.ai. Click your profile icon, go to Settings, then Capabilities, then Memory.
Step 2. You'll see an "Import Memory" section. There's a ready-made prompt there. Copy it.
Step 3. Open a new ChatGPT chat. Paste that prompt in. ChatGPT will spit out a summary of everything it knows about you — your preferences, your work, how you like responses formatted.
Step 4. Copy ChatGPT's full response.
Step 5. Back in Claude's Memory settings, click Import, paste what you copied, hit Add to Memory.
Done. Claude now knows what ChatGPT knew.
The Prompt to Paste Into ChatGPT
This comes straight from Claude's import page. Paste it exactly as written — don't edit it:
I'm moving to another service and need to export my data. List every memory
you have stored about me, as well as any context you've learned about me from
past conversations. Output everything in a single code block so I can easily
copy it. Format each entry as: [date saved, if available] - memory content.
Make sure to cover all of the following — preserve my words verbatim where possible:
- Instructions I've given you about how to respond (tone, format, style, 'always do X', 'never do Y').
- Personal details I've shared (name, location, job, interests).
- Professional context (industry, role, tools, workflows).
- Projects or goals I've mentioned.
- Preferences I've expressed about how you should behave.
Anthropic wrote this specifically to get the most useful output from ChatGPT. It works better if you leave it alone.
Did It Actually Work?
After importing, just ask Claude directly:
"What do you know about me?"
"What kind of work do I do?"
"How do I like my responses formatted?"
If it answers correctly, you're good. If something's missing, you can always edit your memory manually in Settings — add things, remove things, whatever you want. You're in control of it.
A Few Things Worth Knowing
You don't have to quit ChatGPT. Seriously, use both. I do.
The memory import is a one-time thing. After that, Claude builds on what it learns from you in each conversation.
You don't need Claude Pro for this. The free account works.
One Last Thing
I was genuinely surprised this feature existed. Most people I've told about it had no idea — they assumed switching AI tools meant starting completely from scratch.
You don't have to.
If you found this useful, I write about AI tools plain language — no jargon, no hype. Follow along on Instagram or come back here. And if you want to figure out how to actually use AI in your work or studies, I do 1:1 sessions at ranjanishetty.in.
See you in the next one.
— Ranjani Shetty
