Google Removes 'No Data Sent' Claim From Chrome AI Settings

Change sparks debate over how Google handles user data

Google has removed language from Chrome that explicitly stated users' data would not be sent to the company's servers when using certain AI-powered features.

The change appeared in the latest rollout of Chrome version 148.0.7778.97, where wording in the browser's settings menu for "On-device AI" was quietly altered.

Previously in Chrome version 147, the browser's "On-device AI" section stated: "To power features like scam detection, Chrome can use AI models that run directly on your device without sending your data to Google servers. When this is off, these features might not work."

However, the phrase "without sending your data to Google servers" has now disappeared.

The modification was first noticed by users on Reddit before being highlighted by developers and privacy advocates online. Google insists the update does not signal any change in how Chrome processes AI data.

"This doesn't reflect a change to how we handle on-device AI for Chrome," a company spokesperson said. "The data that is passed to the model is processed solely on device."

However, the change in the wording has fueled speculation among critics about why the company chose to revise the description. Alexander Hanff, a computer scientist and lawyer who has previously criticized Google's handling of AI features in Chrome, said the decision raised "serious questions" about transparency.

"Why was the sentence… removed from the on-device AI description in Chrome's Settings UI?" Hanff asked.

He said he asked Google directly whether the original wording had been inaccurate and whether legal concerns had played a role in its removal. Google told Hanff the change was intended to avoid confusion.

The company acknowledged that some websites using Gemini Nano – Google's lightweight AI model built into Chrome – may still be able to access information exchanged with the model through developer tools such as the Prompt API. According to Google, any handling of that information would fall under the individual website's privacy policy rather than Google's own systems.

Chrome Silently Downloading 4GB AI Model File

The controversy follows separate reports last week that Chrome had been automatically downloading Gemini Nano onto compatible devices without clearly notifying users.

Researchers said the browser was storing roughly 4GB of AI model files locally in a folder linked to on-device AI features.

Hanff claimed the files appeared on both Windows and MacOS systems without an explicit opt-in process, and that manually deleting them caused Chrome to reinstall the files during the next restart.

Google said Gemini Nano has been available in Chrome since 2024 and powers security features including scam detection. It added that the model operates locally rather than in the cloud and automatically removes itself on devices with limited resources.

The company also said it had recently added settings allowing users to disable and remove the AI model from Chrome entirely.

"We’ve offered Gemini Nano for Chrome since 2024 as a lightweight, on-device model," Google said.

"It powers important security capabilities like scam detection and developer APIs without sending your data to the cloud. While this requires some local space on the desktop to run, the model will automatically uninstall if the device is low on resources. In February, we began rolling out the ability for users to easily turn off and remove the model directly in Chrome settings. Once disabled, the model will no longer download or update."

This article originally appeared on MES Computing’s sister site Computing.