Chrome Quietly Deleted a Key Privacy Promise — Here’s Why It Matters for You

In early May 2026, observant Chrome users and privacy watchdogs noticed something missing from the browser’s settings page. A long-standing privacy guarantee—one that assured users Google would not send personal data to its servers for certain features—had been silently removed. The change, first reported by Decrypt on May 7, correlates with the rollout of new on-device AI capabilities in Chrome. If you rely on Chrome for everyday browsing and care about where your data goes, this shift is worth understanding.

What happened

The promise in question appeared in Chrome’s settings under “Privacy and security.” For years, it stated that Chrome would not send your personal data—including page URLs, text you enter, or browsing history—to Google unless you explicitly enabled a feature that required it. That language is now gone. Instead, the settings page now describes the new on-device AI features, such as smart tab organization, automatic text suggestions, and enhanced search predictions, as working locally on your device—but no longer includes the blanket assurance that data won’t leave your machine.

Google’s explanation, as reported by multiple outlets, is that the old promise was “no longer necessary” because the new AI models run entirely on your device. The company argues that nothing is sent to its servers unless you opt into cloud-based improvements, and that the on-device processing is inherently private. Critics point out that the removal of the explicit promise creates ambiguity and could allow future features to send data without the same clear user consent.

Why it matters for your privacy

On-device AI, as the name suggests, processes data locally using your computer’s own hardware, rather than sending it to a remote server. That is generally a good thing for privacy—your keystrokes, visited pages, and other input never leave your machine. However, the devil is in the details. Many on-device AI systems still send anonymized or aggregated data back to the developer to improve the model. Google has not explicitly clarified whether such data will be collected from Chrome’s on-device AI, or under what conditions.

The removal of the old promise matters because it erodes the baseline guarantee users once had. Before, you could be reasonably certain that Chrome would not send personal data for any feature unless you turned it on. Now that assurance has been replaced with a more nuanced claim that the AI runs locally, which may be technically true but leaves room for future adjustments. Privacy advocates worry this sets a precedent: if a feature later requires cloud assistance, Google could update the UI without restoring the old language.

What you can do about it

You don’t have to accept the change without action. Chrome still allows you to control many AI-related features. Here are practical steps to limit what the browser can do with your data:

  1. Open Chrome Settings – Click the three-dot menu (top-right), then go to Settings.
  2. Find AI features – Look under the “Privacy and security” section. In recent Chrome versions, there is a subsection labeled “AI and machine learning” or similar. This is where you can toggle individual features.
  3. Disable what you don’t need – Turn off options like “Smart tab search,” “Automatic text predictions,” or “Enhanced spell check.” If you don’t use these, leaving them off reduces any potential data transmission.
  4. Review your sync settings – Even if AI features are off, Chrome sync still sends your bookmarks, passwords, and history to your Google account. Ensure that sync is set to your comfort level, or disable it entirely.
  5. Consider alternative browsers – If you want a browser with a stronger privacy baseline, Firefox, Brave, or the privacy-focused Chromium-based browsers like Bromite may offer clearer data handling policies.

These steps won’t undo the removal of the promise, but they give you direct control over the new features. The key is to be aware that the default setting may now allow data collection without the old warning.

Looking ahead

Google’s move reflects a broader industry trend: browser companies are embedding AI into their products, and privacy promises that were once straightforward are becoming murkier. On-device AI is not inherently bad—it can improve responsiveness and reduce server costs—but the lack of transparent, user-facing guarantees is troubling. If you use Chrome, now is a good time to check your settings and decide how much trust you want to extend to the browser’s new capabilities.

Sources:

  • Decrypt, “Chrome Deleted Its Own Privacy Promise for Sneaky On-Device AI,” May 7, 2026.
  • Yahoo Tech, coverage of the same story, May 7, 2026.
  • Google Chrome Help documentation for on-device AI features (as of May 2026).