Google's Chrome Now Uses Gmail Data to Fill In Order, Tracking and Flight Details on Forms

Google has begun letting its Chrome browser draw on information in users' Gmail inboxes to complete online forms, the company announced on Wednesday. The feature, called enhanced autofill, is initially available in the US on desktop, Android and iOS.

The tool can suggest order numbers, shipment tracking details and flight bookings when a web form requests them, according to an announcement by Amanda Lim, a product manager on Chrome. It is the latest step in Google's effort to weave its Gemini artificial intelligence assistant into the browser used by most of the world's internet users.

How the feature works

Autofill is a long-standing browser function that saves people from repeatedly typing routine information such as their name and postal address. Enhanced autofill extends that idea by pulling details not from data a user has deliberately stored in the browser, but from the contents of their email.

The capability runs on Gemini through a feature Google calls Personal Intelligence. That system allows Gemini to consult connected Google apps, including Gmail and Google Photos, so that its responses reflect a user's own circumstances rather than generic information.

Access is limited. The rollout covers people who use Gemini in Chrome and have already switched Personal Intelligence on. Google's announcement did not give a timetable for availability outside the US.

The announcement identifies only three categories of information as coming from Gmail: order numbers, tracking information and flight bookings. Decrypt noted that credit card numbers and addresses remain part of Chrome's standard autofill, which relies on details users have saved in the browser, and that Google's post does not say the new tool extracts such data from inboxes.

Google's privacy assurances

Google has framed the feature as something users control. Lim wrote that people can decide in Chrome's settings whether enhanced autofill is active and what it is permitted to access. "As always, sensitive data is encrypted and you remain in control," she added.

Encryption converts data into a form that only authorised parties, which in this case include Google, can read. However, the announcement does not explain how encryption applies to the Gmail content that Chrome reads, nor does it set out precisely what the browser scans in order to find relevant details.

The company has made broader commitments around Personal Intelligence. In March, when it began extending the feature to free users in the US, Google said that Gemini and AI Mode, its AI-driven experience within Search, do not train directly on users' Gmail inboxes or Google Photos libraries.

That pledge has limits. Google says users choose whether to connect their apps and can disconnect them at any time, but it does train on some restricted information, such as specific prompts people enter and the responses the model generates.

A pattern of user concern

The launch arrives against a backdrop of recurring unease over how Google uses personal content. In November, users objected to a Smart Features setting that allowed Gemini to analyse their emails and calendars unless they actively opted out. Google responded at the time that its underlying data-handling practices had not changed.

The company also amended its privacy policy with effect from 16 October. The revision added an exception to its commitment not to use content from Gmail, Drive and Photos for advertising: that material can now inform personalised ads where users opt to share it with other Google services offering such advertising.

According to Google, by default this material plays no part in targeting adverts, and advertisers cannot see what users hold in their Gmail, Drive or Photos accounts. Even so, the change illustrates how the boundaries around inbox data have become more conditional as Google's services become more interconnected.

A separate adjustment has also drawn attention. Chrome removed a line from its settings stating that on-device AI, meaning models that run on a user's own computer rather than on Google's servers, operates without sending data to Google. The company said the revised wording did not reflect any change in how it handles on-device AI.

Why it matters

Chrome's reach gives any change to the browser outsized significance. The browser holds about 65% of the market, so features introduced there can quickly shape what hundreds of millions of people come to expect from software.

Enhanced autofill also fits a wider push by Google to make Chrome an active assistant rather than a passive window onto the web. In January the company gave paying US subscribers an AI browsing agent known as auto browse, which navigates websites and completes forms on a user's behalf.

Taken together, the two features point towards a browser that not only knows what a user has saved, but also understands their purchases, travel and correspondence, and can act on that knowledge. For users, the practical appeal is plain: hunting through an inbox for a booking reference or a parcel tracking code is a familiar chore, and automating it could save time.

For Google, the strategic logic is equally clear. The value of an AI assistant rises with the amount of personal context it can access, and Gmail is one of the richest sources of such context the company controls. Linking that data to the browser strengthens the case for staying inside Google's ecosystem.

Risks and what to watch

The trade-off is the familiar one between convenience and privacy. Each new connection between a user's email and another service widens the set of systems handling sensitive information, and the details of how that information is processed are not always fully disclosed.

The gaps in Google's announcement are likely to attract scrutiny. The company has not explained how encryption protects the inbox content Chrome reads, or what exactly is scanned. Its separate assurances, that Gemini does not train directly on inboxes and that Gmail content is not used for ad targeting by default, each come with qualifications that users may not notice.

The opt-in design offers some reassurance. Enhanced autofill is restricted to those who have already enabled Personal Intelligence, and Google says access can be managed through Chrome settings. That contrasts with the opt-out Smart Features setting that prompted complaints in November.

Several questions will determine how the feature is received:

• Whether Google expands the types of Gmail data used beyond orders, tracking and flights
• When, and under what terms, the feature reaches users outside the US
• Whether the company publishes more detail on encryption and scanning
• How regulators and privacy advocates respond to deeper links between email and browsing

For now, enhanced autofill is a modest feature with limited scope. Its significance lies less in what it does today than in what it signals: a browser increasingly built around an AI assistant with access to the most personal corners of a user's digital life, and a company asking users to trust that the controls it provides are sufficient.