OpenAI has announced a ChatGPT memory improvementwith which it not only improves its ability to remember and use information already mentioned by the user, but also makes this context not limited to the ongoing conversation and even updates it in real time.
OpenAI has changed ChatGPT’s memory architecture so that this ‘chatbot’ can have reference information with each new conversation and always keep in mind user preferencesfor example, if they are vegetarians or do not want a certain person to be mentioned, so that the answers adapt to them.
It is about the third update which improves the memory function, since the function was introduced in 2024. Users then had to ask ChatGPT to remember specific information, although last year the possibility was introduced for it to remember information previously mentioned in the conversation to personalize responses.
The new architecture introduces a more powerful and efficient memory systemwhich provides continuity to the context with synthesized memories, which is no longer limited to the current conversation, but is maintained for new conversations, without having to “present itself from scratch”, as OpenAI explains in a statement.
It also improves the understanding the temporal context of informationFor example, if users have mentioned that they are taking a trip in July, ChatGPT will be able to distinguish between present, future and past tenses, to update the information based on the moment, making it more relevant and useful.
Furthermore, introduce a summary of the information that ChatGPT “remembers”“, so that users can review and modify it at any time, and even specify the topics that the ‘chatbot’ can discuss and when it should do so.
This change to the memory feature is available to Plus and Pro subscription users in the United States. It will also be extended to free account users and Go users, after reaching the processing capacity necessary to scale it and do so more efficiently, although its deployment will take place in the coming weeks.