Meta Platforms announced on Wednesday that it is introducing an “incognito” mode for its AI assistant on WhatsApp, giving users a way to hold private, temporary conversations that the company says it cannot access itself. The new mode is meant to address rising unease among people who confide sensitive details—financial figures, health conditions, personal worries—to AI chatbots and do not want that information stored, reviewed, or fed into the models that power the assistants.
Incognito chats are processed in what Meta described as a “secure environment” and are not saved by default; all messages vanish when a session ends. The option, outlined in a company blog post, follows similar privacy controls already available from rivals: Google’s Gemini chatbot lets users turn off chat history and opt out of allowing their data to be used in training, and OpenAI’s ChatGPT provides comparable choices.
Will Cathcart, Meta’s head of WhatsApp, told reporters that the company built the feature because people are turning to AI with deeply personal questions. “We’re starting 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,” Cathcart said.
The incognito mode includes safety features designed to prevent the chatbot from responding to harmful topics. Cathcart said the system 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 can type questions and receive text replies but cannot upload or generate images. Meta requires age confirmation because it does not permit users under 13 on its platforms.