WhatsApp Launches Incognito Mode for Private AI Conversations

Summary

WhatsApp has introduced a new 'incognito' mode for conversations with its Meta AI chatbot, allowing users to have completely private chats that neither they nor Meta can retrieve or access. The feature addresses growing user concerns about data privacy, particularly for sensitive topics including health, relationships, and finances. When activated, incognito conversations produce no server-side logs, and past chats disappear from the user's view after the session ends.

WhatsApp head Will Cathcart explained that users wanted to interact with AI on personal matters but felt uncomfortable sharing that data with the company. Mark Zuckerberg described it as the "first major AI product where there is no log of your conversations stored on servers." The technology uses security measures equivalent to WhatsApp's end-to-end encryption but operates on a separate system. Initially, the feature supports text-only conversations, with built-in guardrails designed to refuse potentially harmful or illegal requests.

However, cybersecurity experts have raised significant concerns about accountability and potential liability. Prof Alan Woodward from Surrey University warned that without accessible chat logs, there would be no way to investigate whether the AI's responses caused harm—including cases involving wrongful death or suicide. The concern centers on users placing trust in an AI system that could lead them astray on critical health or financial decisions with no retrievable evidence.

The launch comes as Meta invests heavily in AI infrastructure, with $145 billion earmarked for 2026. Analysts note that investors are closely watching whether this spending will generate returns through enhanced advertising and commerce capabilities across Meta's platform empire.

• Feature reflects Meta's broader AI infrastructure strategy; company investing $145 billion in AI for 2026, positioning AI as key growth driver for advertising and commerce