However, over time, researchers developed methods to give AI a sort of short-term memory, allowing it to remember information within a single conversation. This was a big step forward, but it still had limitations. Once the conversation ended, the AI’s memory was wiped clean. The latest feature that functions as a type of long-term memory, like the one being rolled out for Gemini Advanced, is a significant advance.
With it, Gemini can now access and process information from past interactions. This has many potential applications. For example, if you’re using Gemini for a long project, you won’t have to re-explain your goals or previous steps each time. Gemini can simply refer back to earlier parts of the conversation. This could be particularly useful for tasks like coding, writing, or research.
Rolling out starting today, you can ask Gemini to consider your past chats to craft its responses. Easily pick up where you left off or have it summarize a previous topic. You can view, edit, or delete any chats you’ve had with Gemini, and see when it’s used.
Try it in Gemini… pic.twitter.com/oTY12Jl0GF
— Google Gemini App (@GeminiApp) February 13, 2025
This new capability is currently available in English for Gemini Advanced subscribers through the Google One AI Premium Plan on the Gemini web and mobile app. The rollout will continue over the next few weeks, with more languages and Google Workspace Business and Enterprise customers gaining access.
I am personally excited about this development and how much time can potentially be saved. Instead of repeating instructions or providing background information, users can now hopefully rely on Gemini’s memory to maintain context.