Claude now lets you bring over memories from ChatGPT, Copilot and Gemini

Home » Python » Claude now lets you bring over memories from ChatGPT, Copilot and Gemini

Without much fanfare but with a clearly strategic move, Claude has gained the ability to pull in your conversational memory from other major AI assistants. People who have spent months building up long chats in tools like ChatGPT, Copilot or Gemini can now carry that context across instead of starting from zero.

Rather than relying on hidden data transfers between platforms, Anthropic has opted for a transparent, user-driven workflow to migrate conversation history. The process revolves around a short, reusable prompt and a line of code that encapsulates your past interactions, which you then paste into a dedicated memory section in Claude.

How Claude imports memory from ChatGPT, Copilot and Gemini

The new feature is built around a simple but structured flow: you ask the original chatbot to package up your context, then you hand that package over to Claude. Anthropic describes this as part of a longer effort to “teach the AI how you work”, rather than forcing you to re-explain your routines, preferences and long-running projects every time you switch tools.

From a practical point of view, the first step is to use Claude to generate a specialized prompt that you copy into ChatGPT, Copilot, Gemini or another supported assistant. That prompt tells the other system to compile your relevant history — including both the conversational memory and the broader context that has been built up over time.

Once the external chatbot processes that request, it doesn’t return a raw transcript; instead, it responds with a compact line of code that represents your stored information. This code acts as a portable container for your preferences, accumulated project details and recurring topics that have shaped your past conversations.

With that line of code in hand, the user goes back to Claude and pastes it into a specific area inside Claude’s memory interface. Anthropic has created a dedicated section for this cross‑assistant import so that the system knows how to decode, store and interpret the incoming data without mixing it haphazardly with new chats.

After the code is submitted, Claude ingests the imported history and aligns it with your personal style of working. The goal is that, from the very first message in your new Claude session, the assistant already “remembers” long‑standing projects, preferred formats, frequently discussed subjects and other patterns that previously lived only inside ChatGPT, Copilot or Gemini.

Continuity, personalization and a smoother switch between AIs

One of the main selling points of the update is that it lets people pick up their conversations almost exactly where they left off, even if they change assistants in the middle. Instead of patiently rebuilding context with long explanations and repeated instructions, users can treat Claude as a continuation of their existing AI workflow.

The imported memory is not limited to isolated facts. According to Anthropic’s description, Claude absorbs communication styles, common themes and long, multi‑step discussion threads. That means the assistant can adapt to whether you prefer concise answers or detailed breakdowns, recognize which projects are ongoing and recall which resources or documents have already been discussed.

This approach also tackles a longstanding irritation for power users: the friction of starting over every time you experiment with a new model. People who keep large archives of conversations or rely on AI tools for complex, recurring workflows — from technical research to creative planning — can now move across platforms without abandoning that accumulated knowledge.

Interoperability has been a consistent user request as the AI market has filled with overlapping offerings. By enabling import from competitors, Anthropic is acknowledging that most users do not live in a single‑vendor bubble. Instead of locking people in, the company is betting on convenience: if the switch feels seamless, more people might be willing to try Claude seriously.

The feature also has a more subtle benefit: it reduces the psychological barrier to experimenting with alternative assistants. Knowing that your months of guidance and fine‑tuning won’t be wasted makes it less risky to test a different model, which in turn raises the bar for all providers to support smoother migration and data portability.

Why the workflow is manual (and why that matters)

Although the process may sound like a direct platform‑to‑platform sync, Anthropic emphasizes that the transfer is not automatic or hidden. Users are always the ones who initiate the export in the original chatbot, copy the generated code and then paste it into Claude’s memory area.

This design has a few implications. On the one hand, it gives people explicit control over what gets imported. You decide which conversations or which account to use, and you can perform the process only when you are comfortable sharing that history with a new assistant. On the other hand, it avoids silent data flows between companies, which could raise privacy or trust concerns.

Technically, the code snippet produced by the external chatbot acts as a structured representation of your historical context. By relying on a coded format instead of raw copy‑paste text, Claude can parse the content more reliably and integrate it into its own memory system in a predictable way.

There is still some work required from the user — copying prompts, waiting for the export, transferring the code — but Anthropic suggests the whole import can be done in under a minute for many users. The company frames this effort as a one‑time setup that can spare you from repeating months of explanations in future sessions.

The decision to keep the user in the loop also reflects a more general industry trend: people are increasingly sensitive to where their chat histories go. A visible, prompt‑based workflow makes it easier to understand what is being shared, when, and for what purpose, instead of leaving those details buried in terms of service.

Boosting productivity for researchers, professionals and heavy users

Beyond casual chatting, Anthropic is clearly targeting scenarios where memory and long‑term context are central to daily work. The company repeatedly points to roles that depend on consistency and continuity across many sessions with an AI assistant.

For example, researchers managing large volumes of historical data can benefit from carrying over their established threads of analysis when they move from one tool to another. Instead of re‑explaining every study, dataset or citation trail, they can have Claude pick up prior lines of inquiry with minimal friction.

Similarly, professionals running long‑term projects — from software development to marketing campaigns — often rely on chatbots to track requirements, brainstorm ideas and document decisions. Importing memory means these ongoing efforts do not become fragmented simply because the underlying assistant changes.

Power users who interact with AI tools dozens of times a day also gain from this approach. Many have grown used to teaching their assistant a particular workflow, style guide or set of rules. Moving that accumulated “training” into Claude can save them from re‑authoring detailed instructions just to get back to the same baseline behavior.

Anthropic frames the ultimate goal as creating a continuous flow of information across chatbots, so that users don’t have to keep track of what was said to which model at which time. In practical terms, that means fewer repeated prompts, less time spent rebuilding context and a smoother day‑to‑day experience for people who lean heavily on AI to manage their tasks.

Claude’s momentum and the “switch without starting over” message

The timing of this update also lines up with a broader shift in the competitive landscape. Recent download charts on Apple’s App Store have shown Claude climbing to the top spot among productivity apps, overtaking ChatGPT in certain periods. That surge comes amid public scrutiny of OpenAI’s partnership with the U.S. Department of Defense, a move that has sparked debate — and, in some cases, user backlash.

Anthropic appears to be capitalizing on this moment by positioning Claude as an easy alternative for people thinking of switching tools. The company has refreshed its interface and marketing around migration, highlighting that the import experience is smoother and more polished than earlier iterations introduced in late 2024.

On its dedicated onboarding page, the message is direct: “Switch to Claude without starting from scratch”. The company claims that the import process can be completed in under a minute and leans on a familiar sentiment for heavy users: “You’ve spent months teaching another AI how you work; that context shouldn’t vanish just because you want to try something new.”

The slogan “Claude can import what matters” encapsulates the pitch. Rather than promising abstract performance gains, Anthropic is emphasizing the preservation of the hard‑won, practical knowledge embedded in your chat history. The idea is that your first real conversation with Claude should already feel like the hundredth, not like the awkward first day with a new colleague.

At the same time, the company is careful to keep the tone pragmatic. The feature is presented less as a flashy add‑on and more as an answer to a very concrete pain point that has emerged as people rely on AI systems for more serious, sustained work.

Taken together, the new memory import workflow, the improved interface and the recent spike in downloads point to a clear strategy: lower the cost of switching, respect users’ existing data and let performance speak for itself. For anyone who has invested time training ChatGPT, Copilot or Gemini to match their habits, the prospect of bringing that history along makes trying Claude a far less disruptive decision.

All of these changes contribute to a landscape where AI assistants are expected to travel with the user rather than trap them. Claude’s new ability to ingest the memory and context from rival chatbots reflects that expectation, offering a more fluid path for people who want to experiment with different models while keeping their past work, preferences and long‑term projects intact.