OpenAI has just rolled out a new feature called ‘memory’, exclusively for ChatGPT plus users, which allows the AI model to remember information users provide across conversations. First revealed in its testing phase in February, memory “saves you from having to repeat information and makes future conversations more helpful,” according to an OpenAI blog. The post explained, “As you chat with ChatGPT, you can ask it to remember something specific or let it pick up details itself. ChatGPT’s memory will get better the more you use it and you’ll start to notice the improvements over time.” Essentially, ChatGPT will remember details about a user, such as work preferences, prior prompts given by the user and even personal details, should the user provide them. GPTs built by users will also have their own individual memories.
Memory can be turned on or off in the personalisation settings and recorded memories can be seen in the same settings tab where they can even be deleted. Users can also ask ChatGPT to similarly ‘forget’ a memory.
Medianama tested the new feature. First we informed ChatGPT that we had just adopted a new puppy named ‘Riley’. The AI appeared to update its memories without any prompting.
Next, we created a new chat and greeted ChatGPT, who inquired after the dog without any prompts.
One can see recorded memories and delete them in the data personalisation settings.
Since ChatGPT will now be ‘remembering’ information about users, it raises questions about data privacy and security. OpenAI informed users that memories, like all other content provided to ChatGPT, would be used to train their models, although that can be turned off in the ‘Data Controls’ section of the settings menu. One can also start a temporary chat, which won’t appear in user history or record memories. However, the AI startup does not appear to have yet enacted any comprehensive safeguards for memories. The blog states, “Memory brings additional privacy and safety considerations, such as what type of information should be remembered and how it’s used. We’re taking steps to assess and mitigate biases, and steer ChatGPT away from proactively remembering sensitive information, like your health details – unless you explicitly ask it to.”
Also Read:
STAY ON TOP OF TECH NEWS: Our daily newsletter with the top story of the day from MediaNama, delivered to your inbox before 9 AM. Click here to sign up today!
Source: ChatGPT Now Remembers User Information, Here’s How It Works
