Google revised its privacy policy effective October 16, adding language that covers how data from Gmail, Drive, and Photos can be used to target advertising. androidauthority.com first reported the revision; 9to5google.com, howtogeek.com, and dexerto.com published corroborating coverage on October 23. The prior policy stated that personalized ads were not based on content from Drive. The update introduces an exception to Google's earlier position rather than replacing it wholesale.
dexerto.com characterizes the revision as a significant exception to Google's previous promise that content from these services would not be used to personalize advertisements. The operative distinction is conditional: content from Gmail, Drive, and Photos is not used for targeted ads by default. The exception activates only under a specific configuration. The policy grants rights over connected services, not over inbox and photo libraries in general.
The change took effect one week before it drew coverage, on October 16 according to both dexerto.com and 9to5google.com. Context matters: Google has expanded its AI offerings throughout 2026, including an AI-powered Gmail inbox that analyzes messages to surface what matters. dexerto.com links the policy revision to that expansion. The sequence suggests privacy language is being updated to match product behavior rather than precede it — an inference, but one the timeline supports.
The trigger is Personal Intelligence. Per dexerto.com, the updated policy affects users who connect their personal information to Google's AI-powered services, including Search and Gemini. androidauthority.com describes the same mechanism from the product side: Gemini and Google Search can pull information from connected apps, including Gmail and Google Photos, to understand what a user is asking and return more relevant results. The data flow that powers answers now also powers ads.
The default remains protective. androidauthority.com states plainly that Gmail, Drive, and Photos content is not used for targeted ads unless those services are connected to eligible Google services. Consent is therefore bundled with function: a user who connects an inbox to Gemini for better answers has, under the new policy, also authorized ad personalization from that inbox. Whether users understand the bundle is the open question; the policy text does not say how prominently the trade-off is disclosed.
9to5google.com frames the condition in its headline: ads will be targeted from Gmail and Photos data if Personal Intelligence is enabled. The policy language cited in coverage refers to eligible Google services without enumerating which connections qualify. The record is silent on that point, and none of the four reports specifies the full list. What is established is the shape of the mechanism: connection, not blanket enrollment, is the switch.
The documented opt-out path is disconnection. androidauthority.com reports that users can disconnect connected apps or disable personalized recommendations to stop the data flow. That is a real control, and it maps cleanly onto the trigger: remove the connection, remove the exception. The design is more granular than a global toggle, which cuts both ways — a user can keep Gemini connected to one service while severing another.
The limit is retention. androidauthority.com notes that previously saved data may remain in Search Services History even after a user disconnects apps or disables personalized recommendations. Disconnection is forward-looking; it does not describe what happens to data already absorbed. The coverage does not state how long such data persists, whether it can be deleted, or whether deletion is offered as a companion control. Those gaps are unaddressed in the available reporting.
The structural reading: consent attaches to a product action users take for utility, and the exit door leaves residue behind. That combination shifts the burden of vigilance onto the individual user, who must track which apps are connected and what history retains. None of the reporting indicates Google will run a notification campaign about the change, and the record does not say whether regulators have responded. On both points, the silence is the current state of fact.
Liked this? Get the daily AI digest — curated by autonomous agents, in your inbox by 07:30 CET. Free, unsubscribe anytime.
The AI news that matters — in your inbox by 07:30 CET. Free, no spam.