Google updates privacy settings to use user media for AI training

As of 7 July 2026, users of Google services are navigating a significant, automated shift in how their personal media and activity data are managed. Under a series of updates to privacy settings initiated in June 2026, the company is authorized to retain a wide array of user-generated content—specifically including images, files, and audio and video recordings—to develop and improve its technologies, including generative AI models and safety measures.

This policy change consolidates previously distinct data-tracking mechanisms under a new "Search Services History" framework. The update, which began reaching customers via email in June 2026, is enabled by default for most users. Unless an individual proactively navigates to their account settings to disable it, interactions with platforms such as Google Search, Google Lens, Google Maps, Google Shopping, Google Flights, Google Hotels, Google Translate, and Google News automatically contribute to the company's internal training data sets.

Media additions

Data Collection Scope and Retention

The scope of this collection includes diverse inputs beyond typed text. According to company documentation, saved media comprises:

• Visual input: Images uploaded or captured through Google Lens for search purposes.
• Audio recordings: Voice inputs from Search Live, general voice searches, and speaking practice sessions within Google Translate.
• Files: Documents and other media uploaded across supported services.

Google has clarified that saved media may be retained for up to 4 years to train its AI models, even if a user deletes the original activity from their account history. While the company states it uses filters to remove sensitive personal information before using the data for model development, the retention of this material has drawn criticism from privacy advocates who view the "opt-out by default" approach as a strategy that exploits user inertia.

Managing Account Settings

To verify or modify these settings, users can access their Google Activity Controls page. Within the "Search Services History" section, individuals can disable the "Save Media" setting, which stops the retention of images, audio, and video for AI training without requiring the deletion of standard search history. Alternatively, users may deactivate "Search Services History" entirely, though this may impact the personalization of recommendations across Google products.

Because the rollout is continuing over several months, some users may not yet see these settings reflected in their accounts. Additionally, those who previously disabled Web & App Activity and Search Personalization toggles may find the new settings already reflected as off.

Industry Context and Regulatory Pressure

Industry observers and privacy experts note that this update follows a broader trend where companies rely on user-generated content to bridge gaps in AI training material. According to Thorin Klosowski, a senior security and privacy activist at the Electronic Frontier Foundation, the strategy leverages the built-in inertia of daily-use apps. I think 'opt in' is really asking the bare minimum of these companies Klosowski said via Wired. Asking their users to consciously choose to enable these features is the least they can do.

For everyday users, the cumulative effect of these platform changes often results in what advocates call "user exhaustion." Ben Winters, director of AI and privacy at the Consumer Federation of America, observed that such shifts create an increasing feeling of powerlessness and hopelessness about even trying to protect your data as consumers manage settings across multiple platforms.

What Happens Next

As Google continues its global rollout of these settings, users are advised to monitor their accounts periodically. While Google maintains that these features provide more relevant results and revisit capabilities, the debate over informed consent remains active.