Google is rolling out its new 'Personal Intelligence' feature, designed to deeply personalize AI Mode in Search by connecting it with users' Gmail and Google Photos data. This opt-in functionality, which delivers tailored responses based on individual information, was announced in a blog post by Robby Stein, VP of Product at Google Search. It is currently available to Google AI Pro and AI Ultra subscribers.
Personal Intelligence enhances AI Mode's capabilities by allowing it to reference information from a user's Gmail and Google Photos. This enables the AI to craft highly relevant search responses, effectively 'connecting the dots' across various Google applications to provide results that align with an individual's unique context.
Availability and How to Enable
The feature is being introduced as a Labs experiment, initially for eligible subscribers in the U.S. using English. It is exclusively available for personal Google accounts and does not extend to Workspace business, enterprise, or education users.
To enable Personal Intelligence, users can follow these steps:
- Open Google Search and tap their profile icon.
- Select 'Search personalization.'
- Choose 'Connected Content Apps.'
- Connect their Gmail and Google Photos accounts.
While the Gmail connection may appear under 'Workspace' in the settings, it's important to note that the feature remains unavailable for Workspace accounts. As the rollout progresses over the coming days, subscribers might also receive direct invitations to try Personal Intelligence within AI Mode.
How Personal Intelligence Works
At its core, Personal Intelligence leverages Google's advanced Gemini 3 model to process user queries in conjunction with data from connected accounts. Once activated, AI Mode can draw upon various personal data points, such as email confirmations, travel bookings, and photo memories, to generate highly informed and relevant responses.
Robby Stein provided illustrative examples in the announcement blog post:
- A user planning a trip could search for activities and receive recommendations tailored to their specific hotel bookings found in Gmail and past travel experiences captured in Google Photos.
- Someone looking to buy a coat might get suggestions that consider their preferred brands, upcoming travel destinations gleaned from flight confirmations, and even anticipated weather conditions.
Stein emphasized the seamless integration this feature offers, stating:
“With Personal Intelligence, recommendations don’t just match your interests – they fit seamlessly into your life. You don’t have to constantly explain your preferences or existing plans, it selects recommendations just for you, right from the start.”
The following screenshots further illustrate the feature in action:
Privacy and User Control
Google underscores that user privacy is paramount, emphasizing that connecting Gmail and Google Photos is entirely opt-in. Users retain full control, with the ability to enable or disable these connections at any time.
The company clarifies that AI Mode does not directly train on the content of users' Gmail inboxes or Google Photos libraries. Instead, training is confined to specific prompts within AI Mode and the model's generated responses, which are utilized to enhance functionality over time.
Google also acknowledges that Personal Intelligence, like any advanced AI, may occasionally make errors, such as incorrectly linking unrelated topics or misinterpreting context. Users are empowered to correct these inaccuracies through follow-up prompts or by providing direct feedback using the 'thumbs down' button.
Significance and Implications
This rollout marks the fulfillment of a "personal context" feature that Google first teased at I/O in May 2025. After Google SVP Nick Fox confirmed in December (seven months later) that the feature was still in internal testing without a public timeline, today's launch brings this long-anticipated functionality to users.
For the 75 million daily active users in AI Mode, as reported by Fox in the same December interview, Personal Intelligence could significantly reduce the amount of manual context required to receive highly tailored responses.
The implications for publishers, however, remain to be fully seen. The impact will depend on how this personalization affects the surfacing of content within AI Mode responses. If the system increasingly prioritizes user-specific context over general search results, certain informational queries might be resolved directly within AI Mode, potentially reducing clicks to external websites. Google has not yet provided data on how Personal Intelligence might influence citation patterns or overall traffic flow to publishers.
Currently, the feature is restricted to paid subscribers using personal accounts. Any future expansion to free users or Workspace accounts would undoubtedly broaden its reach and potential impact.
Looking Ahead
As Personal Intelligence rolls out as a Labs feature over the coming days, eligible AI Pro and AI Ultra subscribers in the U.S. will gain automatic access. The industry will be closely watching to see if Google introduces analytics or attribution tools that allow publishers to monitor how these personalized AI Mode responses influence their content's visibility and traffic patterns.
