In a recent TechCrunch article, it was announced that OpenAI’s ChatGPT is rolling out a new memory feature. This feature allows users to instruct the AI to remember or forget specific details, potentially revolutionizing how businesses interact with AI.
The memory controls are currently being tested with a small group of users, with plans for a broader rollout in the future. Users can explicitly tell ChatGPT to remember something, check what it remembers, or turn off its memory altogether. This feature is designed to enhance the relevance of ChatGPT’s responses, as it can carry what it learns from one chat to the next.
Imagine telling ChatGPT to remember that you live in the suburbs and prefer driving over public transit. Or, in a business context, it could remember your preferred tone, voice, and formatting for blog posts. The possibilities are endless.
OpenAI’s GPTs, custom chatbots powered by OpenAI’s models, also have their own memories. For instance, the Books GPT can remember which books you’ve read and your favorite genres. However, these memories are not shared with ChatGPT.
The memory feature can be disabled at any time, and when turned off, ChatGPT and GPTs won’t create or use memories. Users can also view and delete specific memories or clear all memories from the settings menu.
OpenAI is aware of the potential privacy concerns with this feature. It is taking steps to prevent ChatGPT from proactively remembering sensitive information unless explicitly asked to do so. Additionally, a new Temporary Chat feature is being rolled out, allowing users to have a dialogue with a blank slate, where ChatGPT won’t access previous conversations or memories.
This new memory feature could be a game-changer for businesses, offering personalized experiences and more efficient operations. It’s an exciting development in the world of AI, and we can’t wait to see how it evolves.
