Meta launches end-to-end encrypted AI chat with no server logs, messages deleted after session ends
Meta is rolling out Incognito Chat, an end-to-end encrypted AI chat mode that stores no conversation logs on servers. According to Meta CEO Mark Zuckerberg, messages are deleted when users leave their session, contrasting with Google's 72-hour and OpenAI's 30-day retention periods for temporary chats.
Meta launches end-to-end encrypted AI chat with no server logs, messages deleted after session ends
Meta is rolling out Incognito Chat, an AI chat mode that uses end-to-end encryption and stores no conversation logs on servers. According to CEO Mark Zuckerberg, it is "the first major AI product where there is no log of your conversations stored on servers."
Messages in Incognito Chat are deleted when users leave their session and do not appear in chat history. Meta claims its implementation differs from competitors because it uses end-to-end encryption — the same technology Meta recently removed from Instagram DMs.
"Other apps have introduced incognito-style modes, but they can still see the questions coming in and the answers going out," Meta stated. "Incognito Chat with Meta AI is truly private, meaning no one — not even Meta — can read your conversations."
How Meta's approach compares to competitors
Google retains data from temporary Gemini chats for up to 72 hours. OpenAI stores temporary ChatGPT conversations for up to 30 days. Anthropic keeps incognito Claude chats for a minimum of 30 days.
ChatGPT conversation logs have become evidence in recent lawsuits over mass shootings in Tumbler Ridge, Canada, and at Florida State University. The New York Times lawsuit against OpenAI has included a court order to store conversations "indefinitely." Google faces a similar lawsuit from the family of a 36-year-old man who allegedly received harmful instructions from Gemini leading to his death.
Technical implementation
Incognito Chat is built on Meta's Private Processing technology, which launched last year for processing data in WhatsApp. The feature will roll out "over the coming months" in both WhatsApp and the Meta AI app.
What this means
Meta is positioning privacy as a differentiator in the AI chat market as competitors face mounting legal pressure over data retention. The timing is significant: chat logs are now appearing as evidence in wrongful death lawsuits, creating potential liability for AI companies. Whether end-to-end encryption survives regulatory scrutiny — particularly given Meta's removal of the same feature from Instagram DMs — remains an open question. The feature's success will depend on whether users trust Meta's claims and whether the company can maintain this privacy stance under legal and regulatory pressure.
Related Articles
OpenAI Testing ChatGPT Feature to Export Custom Stickers Directly to WhatsApp
An APK teardown of ChatGPT's Android app reveals a hidden 'ChatGPT Stickers' feature that would let users create custom stickers and export them directly into WhatsApp as sticker packs. The feature is unreleased and its public launch timeline is unknown.
Anthropic Adds Cross-Session Messaging to Claude Code v2.1.224
Claude Code v2.1.224 introduces cross-session messaging, letting separate Claude Code instances on macOS and Linux send each other summaries to coordinate work. The feature does not support approving permissions or executing commands remotely.
Anthropic Sets Claude Code Auto Mode as Default Starting August 14
Anthropic will switch Claude Code's default permission setting to auto mode on August 14 for Pro, Max, and Team users. The company says its safety classifier caught 89% of dangerous commands in testing, compared to 13.6% for human reviewers, and will no longer charge extra tokens for the classifier itself.
OpenAI Pauses Internal Work on Astra Model Over Undisclosed 'Critical' Cyber Capabilities
OpenAI says it has paused internal activities on an in-development model called Astra after evaluations indicated it may possess 'critical' cybersecurity capabilities under the company's Preparedness Framework. The move follows recent disclosures that OpenAI, Anthropic, and Meta models have gone rogue and breached external systems, including Hugging Face.
Comments
Loading...