![Google Gemini AI on a smartphone, integrating Gmail, Photos, and Search for personalized results. [fortune.com]](https://cms.aidailypost.com/uploads/google_search_ai_mode_uses_gmail_photos_new_personal_intelligence_d3ffd0d474.webp "Google Gemini AI on a smartphone, integrating Gmail, Photos, and Search for personalized results. [fortune.com]")
Editorial illustration for Google Search AI Mode uses Gmail, Photos in new Personal Intelligence update
Gemini AI Now Scans Gmail and Photos Personally
Google Search AI Mode uses Gmail, Photos in new Personal Intelligence update
Google is turning your digital life into a concierge. Its latest AI Mode update, rolling out today in the US, can now tap into your Gmail inbox and Google Photos library to personalize search results. The catch?
You have to opt in. For subscribers of Google’s AI Pro and Ultra tiers, this feature, called Personal Intelligence, promises travel itineraries based on actual hotel bookings and a recommendation for that ice cream parlor you’ll love , because, yes, it noticed all those selfies. The company insists privacy is built in, using Gemini 3 without training directly on your data.
But internal testing admits mistakes happen. You can thumbs-down any misfire.
The opt-in Personal Intelligence update aims to automatically tailor AI Mode responses to your interests. For example, Google says Personal Intelligence allows AI Mode to suggest travel itineraries by referencing hotel bookings in your Gmail and holiday snaps in Photos. If you take a lot of selfies with ice cream, then it could curate recommendations for ice cream parlors.
When using AI Mode to shop online, it can prioritize products from brands you wear or have purchased before. Eligible Google AI Pro and AI Ultra subscribers can opt in to the Gmail and Google Photos connections for AI Mode starting today. The update is rolling out in the US as a feature in Google Labs and is only available for personal Google accounts.
This rollout follows Personal Intelligence being introduced to Gemini AI, which can also tap into Search and YouTube history alongside Gmail and Photos. "Built with privacy in mind, AI Mode uses our most intelligent model, Gemini 3, and doesn't train directly on your Gmail inbox or Google Photos library," Google Search VP Robby Stein said in the announcement. "Training is contained to limited info, like specific prompts in AI Mode and the model's responses, to improve functionality over time." Stein says that while Personal Intelligence can be helpful, internal testing shows that "mistakes can happen." AI Mode users can correct any unsuitable recommendations and provide feedback by giving a follow-up response or giving it a "thumbs down."
Your data, your itinerary, your ice cream cravings, Google is now weaving it all into a single, eerily intuitive thread. The promise is undeniable: a search that actually knows you, that surfaces what you need before you’ve fully articulated it. But a tool that reads your inbox and scrolls your photo library in real time demands a new kind of trust.
Google insists privacy is baked in, that Gemini 3 isn’t training on your actual emails or albums. Yet the line between personalization and surveillance grows razor-thin. The opt-in gate is a useful first step, but the real test lies in what happens when the model gets it wrong, and it will.
A mistaken suggestion for an ice cream parlor you hated, a flight itinerary that misreads a hotel confirmation. These aren’t bugs; they’re inevitabilities. The thumbs-down button and follow-up corrections are safety valves, but they place the burden of vigilance on you.
As AI Mode begins to mirror your life, the question isn’t whether it can, it can. The question is whether you’re ready to see yourself reflected in a machine that keeps learning.
Related Reading
Browse all Google news →
LLMs & Generative AI ChatGPT's 'Nerdy' tweak rewards goblin metaphors in answers, study finds
LLMs & Generative AI Google tests visual 'magazine-style' UI for Gemini 3 Pro users
LLMs & Generative AI AI Engineers Face Rising Costs, Need New Strategies for Efficiency
Related Topic NVIDIA and Google Cloud let developers scale AI from prototype to production
Market Trends
Related Topic Google's FACTS benchmark shows 70% factuality ceiling across four tests
Research & Benchmarks
From Archives 2026 Marks Shift of Adaptive GNN‑LLM Integration from Labs to Enterprise
From Archives Anthropic unveils Claude constitution urging builders to ensure safety
From Archives Google hires Hume AI staff to add voice, emotion to DeepMind models
Business & Startups
From Archives India aims for USD 150 bn AI spend by 2026; Google, Microsoft, Amazon commit USD 70 bn
Business & Startups
Common Questions Answered
How does Google's new Personal Intelligence feature work with Gemini?
Personal Intelligence allows Gemini to connect with and reason across Google apps like Gmail, Photos, Search, and YouTube to provide more tailored responses. The feature is initially available only for paid AI Pro and AI Ultra subscribers, and users can individually choose which data sources to connect.
What specific examples demonstrate Gemini's Personal Intelligence capabilities?
Google provided examples like referencing road trip photos to suggest tire recommendations, pulling license plate numbers from images, and helping plan family vacations by analyzing past travel patterns and interests stored across Gmail and Photos. The AI can retrieve specific details from emails or photos to craft more personalized answers.
Is Personal Intelligence enabled by default, and how can users control their data?
Personal Intelligence is off by default and entirely optional for users. Individuals can choose to connect or disconnect specific Google services, use temporary chats without personal data, and Gemini will always cite when it uses personal information. The feature is designed to avoid making proactive assumptions about sensitive data like health information.
