How to Give ChatGPT and Claude the Same Memory

ChatGPT and Claude don't share memory. Each keeps what it learns about you inside its own app, so when you switch, you start over as a stranger. You can copy your memory across once with Claude's import tool, keep one "about me" file that you use in both, or connect both apps to a single outside memory so they read and write the same notes.

Smart, but not about you

Both assistants are smart. What they're missing is you: your work, your plans, your prices, how you like things done. Using two of them can feel like calling a support line twice. The person who picks up is capable, and you still explain everything from the start.

ChatGPT has its own memory. OpenAI describes two parts: saved memories, which are details you ask it to remember or that it saves as useful context, and a setting that lets it draw on your past chats. Claude has a separate memory of its own, which you manage in Claude's settings.

Neither app can see the other's memory. So what you tell ChatGPT on Monday, Claude doesn't know on Tuesday. There are three ways around this. They differ in cost, effort, and how well they stay current.

Option 1: Copy your memory across once

This fits if you're moving from one assistant to the other, or you want to give the second one a head start.

To copy from ChatGPT to Claude, Anthropic's instructions are:

In Claude, open Settings, then Memory, and choose "Start import."
Copy the prompt Claude shows you and paste it into ChatGPT. It asks ChatGPT to list what it has stored about you.
Copy ChatGPT's answer, paste it into the text box in Claude, and click "Add to memory."

To go the other way, ask Claude: "Write out your memories of me verbatim." Paste the result into ChatGPT and ask it to remember the parts that matter.

The limits are real:

It's a one-time copy, not a sync. From the next day on, the two memories drift apart again.
Anthropic calls memory import experimental and says Claude may not always take in what you import.
Anthropic also says Claude may not keep imported personal details that have nothing to do with work.

After an import, open the memory settings and read what landed. Fix anything that's wrong.

Option 2: Keep one "about me" file

This is free and works with any AI tool.

Write one page about yourself. Cover who you are, what you're working on, how you like answers written, and anything the AI should never do.
Save it somewhere you can reach fast, such as a notes app.
Paste it at the start of a new chat. Or put it in the standing instructions each app offers, so you don't have to paste it every time.
When something changes, update the file first, then the copies.

The limit is that you are the sync. Neither assistant can add to the file. If you learn something useful in a long chat, you have to write it down yourself. Stop tending the file and it goes stale, and a stale file is worse than none, because the AI trusts it.

Option 3: Connect both to one shared memory

This is the only option that stays in sync without you doing the work.

Both apps can connect to outside tools through an open standard called the Model Context Protocol, or MCP. A memory service that speaks MCP sits outside both apps. ChatGPT and Claude each connect to it, and each can read from it and write to it. Tell one of them something, and the other can look it up.

Pick a memory service that offers a remote MCP server. Before you sign up, find out who holds your data, whether you can see and fix what it keeps, whether you can export it, and what it costs.
In Claude, go to Customize, then Connectors. Click "+ Add," then "Add custom connector." Paste the server address and sign in.
In ChatGPT on the web, turn on Developer mode in settings. Then create an app for the same server address and sign in.
Test it. In one app, state a fact and ask the assistant to save it. Open a new chat in the other app and ask for that fact.

The limits:

Your plan matters. Anthropic says custom connectors work on Free, Pro, Max, Team and Enterprise, with Free limited to one. OpenAI's developer guide lists Developer mode for Plus, Pro, Business, Enterprise and Education accounts.
You're trusting a third party with your memory. Anthropic warns people to connect only to servers they trust.
The assistant decides when to look things up. Sometimes you'll need to say "check my notes."

Penny’s save and recall widgets show the same session details in two connected assistants. Demonstration with sample data.

Which one should you pick

If you're switching for good, use option 1. If you use both now and then and don't want to pay for anything, use option 2. If you use both most weeks and are tired of repeating yourself, option 3 is the one that lasts.

Where Penny fits

Penny is a memory of the option 3 kind, built so there's nothing for you to tend. Your AI writes notes as it works with you. Penny keeps different kinds of things as what they are: notes, to-dos, the things you track, and how you like things done. When your AI needs a detail, Penny's search finds it and brings it into the chat.

The same memory works in Claude, ChatGPT, Perplexity, Cursor, Codex, and Claude Code, so a price you update in one tool is the price the next one finds.

You can read, fix or delete anything Penny keeps, and export everything from your dashboard anytime: a complete JSON file, your notes as Markdown and your trackers as CSV. Penny holds the encryption keys, because your AI needs to read memory in plain text to use it. Every time Penny's service decrypts your memory, it's on record in your dashboard and sealed into a public, tamper-evident log.

Free to start: 14 days, no card. After that it's $59 a year for founding members, locked for life, or $99 a year on the standard plan. Connecting Claude or ChatGPT requires a paid plan with them.

Frequently asked questions

Does Claude's memory import keep ChatGPT and Claude in sync? No. It's a one-time copy. Anything either assistant learns afterward stays in that app unless you copy it again.

Do I need a paid plan to connect a shared memory? For Claude, Anthropic says Free accounts can add one custom connector. For ChatGPT, OpenAI's developer guide lists Developer mode for Plus, Pro, Business, Enterprise and Education accounts, so a free ChatGPT account isn't on that list.

Is it safe to connect a memory service to my AI apps? It depends on the service. Connect only to one you trust. Ask who holds the encryption keys, who can read your data, whether you can see and fix what it keeps, and whether you can export and delete it.