Meta Platforms has announced a new “incognito” mode for WhatsApp users designed to make conversations with its AI chatbot more private, as concerns continue to grow around how personal data is handled in generative AI systems.
The company revealed in a blog post today that the feature allows users to engage in temporary, private chats with Meta AI, its built-in assistant on WhatsApp. According to Meta, these conversations will not be saved by default and will disappear once a user exits the session.
Meta also said messages in incognito mode will be processed in a “secure environment” that even the company itself cannot access.
The move comes amid ongoing criticism of generative AI tools, which rely on large datasets and have raised concerns about the potential exposure of sensitive user information shared during chats.
Meta said users often turn to AI assistants with personal topics, including financial, health, work, and relationship-related questions, prompting the need for stronger privacy controls.
“We’re starting to ask a lot of meaningful questions about our lives with AI systems, and it doesn’t always feel like you should have to share the information behind those questions with the companies that run those AI systems,” said Will Cathcart, Meta’s head of WhatsApp.
Cathcart also explained that the feature includes built-in safeguards. Incognito chats will not support image uploads or generation, and responses will be limited to text only. The system is also designed to block harmful requests, with Cathcart noting it will “steer the user towards helpful information if it can and then refuse (to answer) and eventually even just stop interacting with the user completely.”
Users will also be required to confirm their age before using the feature, as Meta does not allow users under 13 on its platforms.
Meta’s move places it alongside competitors such as Google and OpenAI, which already offer privacy controls like disabling chat history or opting out of data use for model training.
