Google personalization push for Gemini took two visible steps this week. In the Gemini app, the company announced on October 22 that Personal Intelligence, the opt-in setting that lets the assistant read your Gmail, Calendar, Photos, Maps, YouTube and Search activity, now extends deeper into Google Maps: your own Maps search history and the places you have saved are now answer material, not just generic map data.
The practical change shows up in Google own example. Ask for the best spot to eat after your meetings today, and Gemini checks where your last meeting sits in Calendar, then matches restaurants from your saved places nearby. Previously the Search services connection gave Gemini access to Maps, Shopping, News, Flights and Hotels as data sources; the new step adds the personal layer of searches and saved locations on top. Users opt in through Personal Intelligence settings by connecting Google Search services for Maps and Google Workspace for Calendar.
The same capability is arriving in the car. Google began rolling Personal Intelligence into Android Auto on October 20, letting drivers ask about a flight gate and its delay in one question, or get directions to an appointment pulled from Calendar without dictating the address. Cars with Google built in will follow in the coming months, and Google says personal responses will be shared only while the phone is connected. Photos integration carries extra setup: Face Groups must be enabled and your own face selected, along with the Estimate missing locations control.
The eligibility boundaries are narrow, and worth stating plainly. Personal Intelligence requires users to be at least 18 with a personal Google account, so work, school and supervised accounts are excluded, and the UK, the European Economic Area, Switzerland and Nigeria are not in the launch set. On privacy, Google says Gemini does not train directly on the contents of your Gmail inbox, your Calendar or your Photos library, but it does train on summaries, excerpts and inferences drawn from them, and connected-app data can be reviewed by human reviewers under the Gemini Apps privacy notice.
The second change is a quiet model transition. According to 9to5Google, AI Ultra subscribers are seeing an in-app notice that Gemini 4 Argon will replace Deep Think soon, with a prompt to switch to Argon when deeper reasoning is needed. Deep Think currently exists as a thinking level in Gemini 3.1 Pro, sitting above high, medium and low, and was last updated in February. Google has given no date, and the notice does not clarify whether Gemini 3.1 Pro itself is being retired or only its premium thinking tier.
Read together, the two moves sketch a consolidation: one personalized assistant pulling from every Google surface, powered by one frontier reasoning model, with the older specialty modes folded away. The open questions are the ones Google has not answered: when Argon takes over, what happens to users below the Ultra tier, and whether personalization keeps expanding faster than the regional and account restrictions that currently fence it in.
Comments (0)
Log in to join the discussion
Log InNo comments yet