Google Now Uses Your Drive, Docs, and Media to Train AI by Default

Google has updated its privacy policy, now allowing the company to use data from user accounts — including content from Google Drive, Docs, and media files like images, audio, and video — to train its AI models by default. This change means that unless users actively opt out, their personal and creative assets stored within Google services will contribute directly to the development of Google's AI.

Key takeaways

• Google's updated privacy policy now uses user data from services like Drive and Docs, including images, audio, and video, for AI model training by default.
• This change affects AI-art creators by potentially exposing their stored projects, prompts, and generated media to Google's AI systems.
• Users must manually navigate to their Google Activity Controls and disable the "Web & App Activity" setting to prevent their data from being used for AI training.
• Opting out may reduce the personalization of Google services, but it ensures greater control over sensitive creative assets.

How Your Data Feeds Google's AI

The policy adjustment, which went into effect recently, broadens the scope of data Google collects to improve its AI. Previously, Google primarily used publicly available web data for training. Now, content you create or upload to Google services — from documents in Google Docs to images and video stored in Google Drive or Photos — can be analyzed and incorporated into the models that power tools like Gemini and Google Search's AI Overviews. For AI-art creators, this means that prompts, generated images, reference materials, and even early concept art stored in Google's ecosystem could become part of the training datasets that shape future AI capabilities.

This shift is significant because it moves beyond generic web scraping to leverage proprietary user data. While Google states the aim is to enhance its AI models, including improving their understanding of language, context, and creative expression, the implications for privacy and data ownership are substantial. Creators often store sensitive or unreleased projects in cloud storage, and this default opt-in means those assets are now part of a broader data collection effort.

Opting Out of AI Training

Users who wish to prevent their data from being used for Google's AI training must manually adjust their privacy settings. The primary control point is within Google's "Activity Controls" section, specifically the "Web & App Activity" setting. Disabling this setting stops Google from saving activity from searches, Chrome, and other services to your Google Account, which in turn prevents that data, including associated media, from being used for AI training. According to TechCrunch, this is the main lever to pull.

To opt out:

1. Go to your Google Account settings.
2. Navigate to "Data & privacy."
3. Under "History settings," select "Web & App Activity."
4. Toggle this setting off.

It's important to note that opting out of "Web & App Activity" might reduce the personalization of some Google services. Features that rely on your past interactions, such as tailored search results or predictive text suggestions, may become less effective. However, for creators prioritizing privacy and control over their digital assets, this trade-off may be acceptable. The decision hinges on balancing convenience against the desire to keep creative works and personal data separate from AI training pipelines. This change underscores a growing trend where user consent for data usage is subtly shifted, placing the onus on individuals to manage their digital footprint actively.