On Thursday, April 10, OpenAI began rolling out a new memory feature in ChatGPT designed to personalise responses based on a user’s past conversations.

Listed in the chatbot’s settings as 'reference saved memories,' the feature aims to enhance the relevance of interactions. According to OpenAI, the update will allow ChatGPT to incorporate past context across its text, voice, and image generation capabilities.

The rollout will initially be limited to ChatGPT Pro and Plus users, excluding those in the U.K., EU, Iceland, Liechtenstein, Norway, and Switzerland. OpenAI explained that additional regulatory reviews are required in these regions.There’s currently no timeline for when the feature will be available to free-tier users.

With this memory feature, OpenAI aims to make ChatGPT more natural and personalised. Users can disable memory entirely or manage specific saved memories through the settings. Users can also ask ChatGPT what it remembers or use Temporary Chat mode to ensure nothing gets stored from the conversation.

While last year’s update allowed ChatGPT to remember or forget information upon request, it relied on users prompting the bot directly. This new rollout aims to make memory management more seamless and intuitive.