OpenAI rolls out ChatGPT memory to select users

OpenAI has begun rolling out memory capabilities to a select number of ChatGPT users this week. Memory will allow the conversational agent to recall details from previous chats in order to provide more personalised and contextually relevant responses. 

According to OpenAI, ChatGPT’s memory will improve over time as users interact with it more. For example, ChatGPT could remember a user’s preferred meeting note format, tailor birthday card suggestions based on a user mentioning their child’s interests, or apply a user’s chosen tone and voice when drafting blog posts without needing to be told repeatedly.

Users remain in control of ChatGPT’s memory and can turn it off completely if desired. They can also selectively delete memories or clear all memories through ChatGPT’s settings. OpenAI states that memories evolve based on interactions and are not tied to specific conversations.

For enterprises and teams, OpenAI believes memory will lead to more efficient workflows by retaining preferences, styles, and past context. Account owners can disable memory for their entire organisation if needed.  

ChatGPT’s memory capabilities will also extend to third-party Generative Pre-trained Transformers (GPTs) built on top of the platform. Developers can choose whether to enable memory for their GPT bots.

OpenAI says they are assessing potential biases with memory and aim to avoid ChatGPT proactively remembering sensitive user details without explicit permission. Additional privacy and safety considerations are being evaluated during this early preview period.

The company plans to share more details around a broader rollout of ChatGPT’s memory capabilities soon. For now, only a small portion of free and paying ChatGPT users will have access.