Google has quietly changed its privacy settings in a way that allows the company to store significantly more of your personal data, including images, files, and audio and video recordings, to improve its AI models. According to TechCrunch, the update means that if you upload any media to Google's Search services, it can now be used to train the company's artificial intelligence unless you manually opt out. The change has renewed scrutiny of how tech giants gather data for AI, and it's the kind of story our breaking AI news coverage tracks closely as data practices evolve.

An Opt-In Disguised as Choice

The shift came through an under-the-radar update to Google's Search services privacy settings, first announced in June via a customer email. As TechCrunch's Sarah Perez reported, the company essentially opted people into expanded AI training while framing the change as giving users more control over their saved history and personalized recommendations.

The update introduced two new settings: Search Services History and Personalized Recommendations. Together, they govern how user activity is used to personalize the Google experience and how long web and app activity is retained. Rather than asking users to affirmatively consent to AI training, Google turned the new data collection on by default, requiring users to find and disable it themselves.

Privacy advocates have criticized the design as a classic "dark pattern," where the path of least resistance leads users to share more data than they may realize.

What's Now Being Captured

The expanded training scope goes well beyond text search queries. According to TechCrunch, the change applies across Google's broader search ecosystem, including:

  • Google Maps
  • Shopping
  • Flights
  • Hotels
  • Translate
  • News

That breadth means media flowing through many everyday Google products can now feed the company's AI models. The update specifically covers images, files, and audio and video recordings.

Concrete examples

Several common interactions now fall under the umbrella. When you use Google Lens to search for something visually by snapping a photo, that image may be saved for AI training. If you use the newer Search Live feature to search via voice input in the Google app, those audio recordings can be stored. And if you use Google Translate to practice speaking a language, that audio is saved as well.

For users who assumed their voice searches or photo lookups were transient, the expansion marks a meaningful change in what Google retains and for what purpose.

Why Google Wants the Data

The shift reflects a broader industry trend toward gathering training data by any means necessary. As TechCrunch noted, rather than relying solely on information scraped from the open web, Google and its competitors increasingly want first-party data captured directly from their own products.

High-quality, multimodal data, images paired with context, audio with transcripts, and files with metadata, has become one of the most valuable resources for improving AI models. For Google, which competes with OpenAI, Anthropic, and others across text, image, and voice AI, its own services are a uniquely rich source of such data. The privacy change effectively turns billions of everyday user interactions into potential training material.

The move also comes as web publishers have grown more aggressive about blocking AI crawlers, pushing companies to find alternative data sources.

How to Opt Out

Users concerned about the change can disable the new collection through their Google account settings. The relevant toggles are the Search Services History and Personalized Recommendations settings, which control how activity from Google's search services is stored and used.

Steps to limit data collection

1. Open your Google Account privacy settings.
2. Locate the Search Services History setting and turn it off to stop media from being saved for training.
3. Review Personalized Recommendations, which also affects how your activity informs personalization.
4. Check Web & App Activity controls to adjust how long your data is retained.

Disabling these settings reduces the data Google stores from your search services interactions, though it may also limit some personalization features.

The Privacy Backlash

The update has prompted fresh criticism from privacy researchers and digital rights groups, who argue that AI training should require explicit, informed consent rather than default enrollment. The core complaint is structural: by turning data collection on automatically, Google shifts the burden onto users to discover and disable it, knowing that most will never change default settings.

Google has defended the change as part of giving users more transparency and control over how their activity is used. The company says the settings are clearly explained and that users can adjust them at any time.

A Wider Industry Pattern

Google is hardly alone. Meta, X, and other platforms have similarly updated terms to permit AI training on user content, often with opt-out rather than opt-in mechanisms. As the scramble for training data intensifies, default-enrolled data collection is becoming the norm rather than the exception across the consumer internet.

Regulators in the European Union have already signaled scrutiny of such practices under the GDPR, which generally requires explicit consent for processing personal data. How U.S. regulators respond remains an open question, though state privacy laws in California and elsewhere may eventually constrain default data collection.

For now, the practical takeaway for users is simple: if you use Google, you may be training its AI, and the only way to stop it is to opt out yourself.

Stay Ahead of AI

For ongoing coverage of AI privacy, data ethics, and how the biggest tech companies are reshaping their data practices, follow AI Buzz Wire for the AI industry coverage that keeps you informed.

Read more AI ethics news →