Google Launches Personal Intelligence in Gemini Using Gmail and Photos
Google is taking its AI assistant to the next level with the launch of a new beta feature called Personal Intelligence in the Gemini app, officially announced on Wednesday, January 14, 2026. This new capability allows Gemini to connect with personal data from Gmail, Photos, Search, and YouTube viewing history to generate responses that are tailored to each user’s real-life context, rather than relying solely on general information from the internet.
What sets Personal Intelligence apart from traditional data retrieval is its ability to perform reasoning across data sources. For example, when planning a trip, Gemini can combine hotel confirmation emails from Gmail with past travel photos stored in Photos to recommend destinations you are more likely to enjoy. It can even recall details such as a car’s license plate number from a photo on your device when you forget it at a repair shop. According to Google, this represents a shift toward proactive intelligence, where the AI understands context without needing explicit instructions on where to find information.
Privacy remains a major concern, and Google has addressed this by making Personal Intelligence off by default. Users must manually enable the feature and choose which apps they want to connect. Google also emphasizes that personal data from Gmail or Photos is not used to directly train its AI models. Instead, the data is referenced only to generate responses in real time. Additional safeguards are in place for sensitive information such as health-related data, which the AI will not infer unless directly prompted.
On the technical side, Personal Intelligence is powered by the latest Gemini 3 model family, featuring support for context windows of up to one million tokens. This is paired with a technique known as Context Packing, which allows the AI to selectively retrieve only the most relevant text, images, or videos for processing. As a result, multimodal data can be handled more efficiently and accurately, enabling smoother and more natural interactions.
Currently, Personal Intelligence is available to Google AI Pro and AI Ultra subscribers in the United States. Google has confirmed plans to expand access to additional countries, including users on the free tier, later this year. There are also indications that this technology will be integrated into Google Search’s AI Mode and may even serve as a foundational element for a next-generation version of Apple’s Siri, following the recently announced collaboration between the two tech giants.
This development highlights a major transition in AI evolution, moving from simple information retrieval toward becoming a true personal assistant that understands users through their digital footprints. For those eligible to try it now, Personal Intelligence can be enabled directly from the settings menu within the Gemini app by selecting which connected apps to allow.
Source: TechCrunch
