A significant structural change in consumer chatbot privacy has taken place over the past two years since Meta launched Incognito Chat with Meta AI on 13 May 2026. As a result of this announcement, the architecture Christakis has been referring to as Sealed Mode in Part 1 of his study on consumer chatbot confidentiality has become a mass-market product and no longer remains a research aspiration. 

The Meta AI app allows WhatsApp users to communicate with the provider in a mode that does not allow Meta to read the conversation, in a similar fashion to the way Meta cannot read two user WhatsApp messages. 

The protection is architectural rather than contractual: Meta has renounced access to content through its hardware design in a Trusted Execution Environment where the chat is processed. Furthermore, the announcement comes as legal and regulatory scrutiny grows on how artificial intelligence providers retain conversational data and respond to law enforcement demands. 

In spite of Google's statement that temporary Gemini chats may be retained for up to 72 hours, OpenAI and Anthropic maintain substantially longer retention periods for temporary and incognito interactions, with ChatGPT sessions and Claude sessions reportedly remaining available for at least 30 days. It has become increasingly necessary to maintain these retention practices since chatbot logs have been used as evidence in numerous high-profile legal cases, including investigations relating to the mass shootings at Tumbler Ridge and Florida State University, as well as a court order requiring indefinite storage of certain ChatGPT conversations in The New York Times litigation. 

Additionally, Google is facing litigation regarding allegations that Gemini encouraged a series of “missions” preceding the death of a 36-year-old man. Meta is positioning Incognito Chat to distinguish itself from conventional cloud AI architectures against this backdrop. Using Meta AI, the company has extended the company's existing Private Processing framework originally deployed within WhatsApp for AI-driven summarization and writing tools directly into conversations with users. This eliminates the previous model of prompts leaving WhatsApp's encrypted channel and reaching Meta's server infrastructure during processing, eliminating the problem. 

Using Incognito Chat, Meta claims that conversations are processed within a Trusted Execution Environment where neither Meta nor WhatsApp has access to plaintext conversation history, while all contextual memory is removed once a session is completed. A web search initiated by Meta AI is also detached from user identity metadata and can be disabled completely by the user at launch. At launch, Meta will provide text-only interactions, with an upcoming "Side Chat" feature that will enable users to privately assist within an active WhatsApp conversation without interrupting the encryption thread. 

Through the new model, Meta AI users will be able to initiate Incognito Chat sessions where they will be able to conduct temporary encrypted interactions. These interactions will be processed in an isolated, secure computing environment whose operations are even inaccessible to Meta AI's internal systems, according to Meta AI. 

By design, Meta says these sessions are ephemeral, with conversations neither being stored nor retained by default following their conclusion. The feature is positioned in a way similar to transient secure messaging rather than conventional cloud-based AI assistance. In the near future, this capability will be available both through WhatsApp and Meta AI's standalone application, along with another privacy-focused feature internally referred to as Sidechat. 

With Sidechat, users will be able to use Meta AI discreetly within an active WhatsApp conversation to summarize exchanges, answer contextual questions, and provide assistance with ongoing conversations without interrupting or exposing the primary encrypted chat thread by invoking Meta AI discreetly within an active conversation. Meta officially stopped supporting end-to-end encrypted direct messages on Instagram less than one week before the rollout, which has increased industry scrutiny.

According to Instagram's support documentation, encrypted direct message functionality will cease on 8 May, and users are advised to export any media or conversations they wish to keep. Users seeking encrypted communication were immediately redirected to WhatsApp, which was explicitly referred to as Meta's sole remaining end-to-end encrypted messaging platform. 

Following the Instagram encryption rollback, a spokesperson from the company indicated that limited adoption prompted the rollback, stating that only a small percentage of users enabled encrypted direct messages, but stressed that WhatsApp's infrastructure could still be used by those who needed encrypted communication.

Meta’s Incognito Chat initiative ultimately represents more than a new privacy feature it signals a broader shift in how major AI platforms are attempting to redesign trust at the infrastructure level rather than through policy language alone. By combining encrypted messaging pathways with Trusted Execution Environment-based processing, Meta is testing whether consumer AI systems can operate with reduced provider visibility while still delivering real-time contextual assistance at scale. 

Yet the rollout also exposes the growing contradiction at the center of the AI industry: as chatbot interactions become increasingly personal, legal demands for data retention, safety monitoring, and platform accountability continue to expand in parallel. Whether Meta’s architecture can withstand both regulatory pressure and public skepticism may determine how future AI communication systems balance usability, privacy, and operational transparency.