WhatsApp has introduced a new 'incognito' mode for its Meta AI chatbot, promising fully private conversations that are invisible to anyone, including the tech company itself. However, the feature has drawn criticism from online safety experts who warn it could make it harder to prevent harm.
What Is the New Incognito Mode?
The incognito mode ensures that all conversations with the Meta AI chatbot are processed in a secure environment that WhatsApp cannot see. By default, messages are not saved, meaning they disappear without leaving a trace. This move comes after users complained about being unable to switch off the previous incognito mode.
Meta, which owns WhatsApp, Instagram, Facebook, and Messenger, said the change addresses privacy concerns. Will Cathcart, head of WhatsApp, stated: '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.'
Safety Features and Limitations
The incognito mode is not a complete free-for-all. Cathcart explained that the chatbot will include safety features to prevent it from responding to harmful topics. It will steer users toward helpful information when possible and may refuse to answer or stop interacting if necessary.
Unlike chatbots such as Claude, Gemini, and ChatGPT, users cannot upload or generate images while using the incognito WhatsApp AI chat. This limitation is designed to reduce potential misuse.
Criticism from Safety Experts
Andy Burrows, chief executive of the Molly Rose Foundation, a charity focused on online safety and suicide prevention, criticized Meta's decision. He called it 'high-risk and deeply irresponsible,' arguing that encrypted, disappearing AI conversations would make it 'impossible' to find evidence if chats contributed to harm.
Concerns about AI and abuse are growing. Companies like OpenAI and Google have faced lawsuits over allegations that chatbots contributed to deaths. Privacy worries around chatbot data also persist.
Despite the backlash, Meta continues to roll out the feature, aiming to balance privacy with safety. The company has not announced any plans to reverse the change.
