This month, Gemini has introduced an innovative feature known as Personal Intelligence, designed to streamline user experience by harnessing the vast amount of information Google has accumulated about its users. This feature is entirely opt-in, allowing individuals to customize the specific data Gemini can access from various Google services, including Gmail and Google Photos. For those subscribed to Google AI Pro or Ultra, a prompt titled “Supercharge Gemini with Personal Intelligence” has likely appeared in the app recently, inviting users to explore this new functionality.
Launched on January 14, Personal Intelligence has prompted users multiple times through pop-up notifications across different devices. When faced with this prompt, users can choose to either select Not now to defer the feature or click Get Started to begin utilizing it. Importantly, users can modify their choices at any time, leading to a pertinent question: should the average user embrace Personal Intelligence or remain cautious?
Initially, I believed I had disabled Personal Intelligence. However, I soon discovered that it is an evolution of Gemini’s previous feature, Personal Context, which is enabled by default. This means that even if you opt out of Personal Intelligence, certain aspects remain active. Personal Context previously allowed Gemini to learn from past interactions, tailoring future conversations based on user preferences and history. This learning process was based on users’ Gemini Apps Activity, which could be managed and deleted at will.
Understanding the Transition to Personal Intelligence
Personal Intelligence now serves as the overarching term for Gemini’s personalization features, effectively replacing Personal Context. The memory-based functionalities that were once categorized under Personal Context are now referred to as Past Gemini chats. This feature retains its default activation status, even if users select Not now for Personal Intelligence.
Previously, users could instruct Gemini on how to respond in various scenarios, providing specific facts for it to remember. This capability continues under the new Instructions for Gemini section within Personal Intelligence. The key distinction with the updated feature is that it personalizes user experience without requiring users to input information actively.
New functionalities can be accessed under the Connected Apps section in the Personal Intelligence tab. By linking Google apps to Gemini, users can opt to allow the AI to utilize insights from these connections to enhance their experience. For instance, if a user connects Gemini to Google Photos and requests information such as their license plate number, the AI can search through the photo library to provide an answer.
This integration spans various Google Workspace applications, including Gmail, Drive, Photos, Home, Search, and YouTube. Each app connection features toggles, enabling users to select which data sources contribute to Personal Intelligence. Google assures users that only their prompts, not personal information from connected apps, are utilized for training AI models.
While some may feel hesitant about sharing more information with AI, Google presents a persuasive argument in favor of enabling Personal Intelligence. The company emphasizes that privacy is paramount, stating that app connections are off by default, and users have complete control over which apps to link. Furthermore, Google maintains that sensitive data remains secure within its ecosystem, alleviating concerns about external data sharing.
We built Personal Intelligence with privacy at the center. Connecting your apps is off by default: you choose to turn it on, decide exactly which apps to connect, and can turn it off anytime. When enabled, Gemini accesses your data to answer your specific requests and to do things for you. And because this data already lives at Google securely, you don’t have to send sensitive data elsewhere to start personalizing your experience.
In essence, Google argues that since users already trust the company with their personal data, allowing Gemini to access connected apps does not introduce additional privacy risks. This perspective is compelling, considering that many willingly share their data with tech giants. However, the decision to enable Personal Intelligence ultimately rests with each user.
Personally, I have opted to keep Personal Intelligence disabled while continuing to utilize Personal Context and Instructions for Gemini. This approach allows me to selectively provide information when necessary, minimizing unnecessary data sharing. I find that I can effectively manage my inquiries without granting the AI access to all my information. For example, I can upload documents to Gemini when needed or use specific tools within Google Photos to retrieve information like my license plate number.
The minor advantage of consolidating features under the Gemini app does not outweigh my preference for maintaining control over my data. Ultimately, the choice is a personal one, and each user must weigh the convenience of enhanced features against potential privacy concerns.
