Google appears to be developing a new feature for its Gemini AI model called the Gemini Personalization model, as revealed through an APK teardown of the latest Google app beta (version 16.8.31). This experimental model is designed to provide personalized responses by accessing a user's Google Search history. The feature aims to refine the chatbot's ability to deliver contextually relevant answers based on individual search patterns and interests.
Google to release a new experimental Gemini model called "Personalization"
— AssembleDebug (Shiv) (@AssembleDebug) March 7, 2025
The Gemini Personalization model will access your Google Search history to offer more relevant responses.
✅ Details and more screenshots - https://t.co/CuGlnHAQZX#AI #Gemini pic.twitter.com/5P13nPLaub
The Gemini Personalization model includes functionality that allows users to opt-in for the AI to utilize their Google Search history. Upon activation, users must have their Web & App Activity setting enabled. The chatbot will then analyze recent search activity, offering tailored summaries and insights. For example, during testing, the model provided a comprehensive overview of searches conducted over the past two months. A disclaimer on the chat screen emphasizes transparency by informing users that their search history is being accessed, with an option to disconnect the feature at any time. While promising, this feature is still experimental and may not function flawlessly upon release.
About Google
Google's Gemini AI represents its ongoing efforts to integrate generative AI into its ecosystem, competing with other advanced models like OpenAI's GPT-4o. The personalization aspect aligns with Google's broader strategy of leveraging user data to enhance AI capabilities while maintaining user control over privacy settings. This development could potentially strengthen Google’s position in the chatbot market by offering responses tailored to individual preferences and needs.
If implemented successfully, this feature could transform how users interact with AI-powered tools by making them more responsive to personal contexts. However, it raises privacy concerns due to its reliance on search history data. Transparency measures like disclaimers and opt-out options may help alleviate these concerns. While no official timeline has been announced, Google reportedly has a Gemini update scheduled soon, which might include this experimental model for Gemini Advanced users.
