Meta says it is rolling out improvements to Meta AI, its cross-platform chatbot, including the ability to make the bot “remember” details of conversations.
In a post on Meta’s official blog, the company said that, in chats with Meta AI on Facebook, Messenger, and WhatsApp for iOS and Android in the US and Canada, users can now tell Meta AI to remember certain things about of them, like that they love to travel and learn new languages.
The memory feature, similar to the memory features of OpenAI’s ChatGPT and Google’s Gemini, allows Meta AI to detect “important details” based on context, according to Meta. For example, if a user mentioned in a previous chat that they are vegan and asks Meta AI for breakfast ideas, the chatbot will constantly take that dietary preference into account.
Meta says that Meta AI won’t remember things in group chats, and that users can delete its memories at any time.
In another, perhaps more controversial, update to Meta AI, Meta says the chatbot will now use account information from all of Meta’s apps to provide personalized recommendations. This information could include the home location on a user’s Facebook profile or recently viewed Instagram videos.
Here’s how Meta CEO Mark Zuckerberg presented it in a post Monday:
“The meta AI will start giving you answers based on the preferences and information you have shared,” he wrote. “For example, he has helped me come up with creative bedtime stories for my daughters, so if I ask him for a new one, he remembers that they love mermaids.”
Personalized recommendations will initially appear on Facebook, Messenger and Instagram in the US and Canada. There won’t be an opt-out option, a Meta spokesperson told TechCrunch.
Meta is doing its best to put a positive spin on Meta AI’s new information scraping capabilities. But considering how few people trust Meta – and Facebook in particular – with their data, one wonders how updates will be received.
