Google Bets on Personalized AI at Scale by Linking Gmail, Photos, and Search
The feature, limited to Google AI Pro and AI Ultra users in the US for now, pulls from users’ own data, tightening the link between AI assistance and behavioral insight.
Topics
News
- Adani Power Sets Up Nuclear Subsidiary
- Musk Unveils xAI Overhaul, Lunar AI Ambitions
- Former GitHub CEO Dohmke Raises $60 Million to Build AI Code Infrastructure
- Leadership Shakeup Deepens at xAI as Two Co Founders Exit
- India Slashes Social Media Takedown Window to Three Hours
- Cisco Moves to Relieve AI Data Center Gridlock With New Chip
Google is beginning a limited rollout of a new feature called Personal Intelligence that lets users link their Google apps to its Gemini artificial intelligence assistant, allowing the AI to reference personal data from Gmail, Google Photos, Search and YouTube to deliver more context-aware responses.
Users must opt in to enable Personal Intelligence, and app connections are off by default. Once enabled, Gemini can pull specific details from connected apps to tailor its answers or suggestions, such as using past photo history or email contents to support planning or recommendations.
Google said it designed the rollout so that individuals have control over which apps are linked and can disable personalization or use temporary chats that do not draw on connected data.
Josh Woodward, Vice-President of Google Labs, Gemini and AI Studio, described the launch on X as a step toward deeper personalization.
“Introducing Personal Intelligence. It’s our answer to a top request: you can now personalize @GeminiApp by connecting your Google apps with a single tap,” Woodward wrote, adding that the beta is limited to Google AI Pro and AI Ultra subscribers in the US.
The company is positioning it as a convenience feature rather than a fundamental shift in how Gemini works. Users can see or ask where information came from, correct responses in real time, or regenerate answers without personalization. Temporary chats are also available for conversations that don’t draw on connected data.
Google is also acknowledging the limits of the system. In its beta form, Gemini can make incorrect assumptions or over-personalize, connecting unrelated data points or missing context, such as mistaking repeated golf-course photos as a personal interest rather than a family obligation.
On privacy, Google said Gemini does not directly train on content from Gmail inboxes or Photos libraries. Instead, those sources are referenced to generate responses, while model training relies on prompts and outputs after steps are taken to filter or obfuscate personal data. Sensitive areas like health are excluded from proactive assumptions, according to the company.
Access to Personal Intelligence is rolling out over the next week to eligible paid users in the US and will work across web, Android, and iOS.
Google said it plans to expand the feature to more countries and eventually to free users, but it will not be available to Workspace, enterprise, or education accounts for now.
