/ai-search515 words

Google's AI Mode in Search gets incognito-style temporary threads

Google's AI Mode now offers temporary threads: one click detaches a conversation from your account, keeping it out of search history and future personalization, on desktop and mobile globally.

Google’s AI feature in Search gets its own incognito mode
Google’s AI feature in Search gets its own incognito modeNik / unsplash
  • Google announced temporary threads for AI Mode in Search via a blog post.
  • Rollout is global, on desktop and mobile, across all supported languages.
  • Temporary threads are activated via an icon in the upper right corner, no settings change required.
  • Sessions detach from the account: no history logging, no effect on future personalization.
  • The feature mirrors temporary chats already available in the standalone Gemini chatbot.

Google has begun rolling out "temporary threads" in AI Mode, its conversational search interface, giving users an incognito-style option that keeps conversations out of search history and stops them from shaping future recommendations. The company confirmed the rollout in a blog post, saying the feature is going out globally on desktop and mobile across all supported languages.

The feature works like temporary chats in standalone AI chatbots. Google's own Gemini app already offered ephemeral conversations, and rivals such as ChatGPT have long shipped similar options. Until now, however, there was no equivalent inside Search itself: if you were dropped into AI Mode — the conversational interface layered onto Google's results — your exchanges were treated like any other search activity.

How the feature works

No settings changes are needed. Users click a temporary-threads icon in the upper right corner of AI Mode to start an ephemeral conversation. Google says the experience "essentially works the same as a temporary chat would in the stand-alone Gemini chatbot."

Once a temporary thread starts, the session detaches from the user's account. Two consequences follow, according to Google:

  • The AI's responses won't be informed by past preferences, since the conversation carries no account context.
  • The thread won't feed future personalization — nothing discussed will resurface in search history or recommendations later.

Closing the thread returns the user to ordinary searching.

What does this change for searchers and publishers?

For users, the practical case is straightforward: anyone digging into a personal or sensitive topic through AI Mode — a health question, a gift research trail, a one-off curiosity — can now do so without Google logging the discussion or recalculating their interest profile around it. Google itself frames a second use case: people who want to learn about something as a one-off without the company starting to treat the topic as a main interest.

For the search industry, the launch is a small but telling data point. AI Mode conversations have, until now, behaved like account-bound search activity. The arrival of a detach mode confirms that these exchanges feed Google's personalization systems in the first place — and that Google now sees enough sensitive, high-volume usage inside AI Mode to warrant an opt-out at launch scale, across all supported languages and both platforms.

A privacy feature, not a ranking change

This is a confirmed product announcement from Google, not a ranking or policy update: nothing here alters how pages are evaluated, indexed, or surfaced in AI Mode responses. Sites and SEO teams monitoring AI-driven referral traffic should not expect ranking movement from it. What it does signal is Google's continued investment in AI Mode as a full conversational destination — one now mature enough to inherit the privacy plumbing of both classic Search (Incognito) and Gemini (temporary chats).

Worth watching next: whether Google publishes usage or adoption figures for temporary threads, and whether ephemeral AI Mode queries eventually appear — or are explicitly excluded from — the query data surfaces, such as Search Console, that publishers rely on to understand AI-driven visibility.

via techcrunch.com (Source)

More from Tom Whitfield

Tom Whitfield

Show full bio

News editor covering regulation, policy and their impact on Search rankings and algorithm updates, generative engine optimization at SERP Journal.

183 articles