Chrome’s latest AI-powered History feature highlights how AI is transforming traffic generation, emphasising the need for SEO to evolve accordingly. In a recent LinkedIn post, a Google Chrome engineer detailed the AI History tool and its method of retrieving signals from previously visited websites. The post suggests that natural language search within browser history could emerge as a new traffic channel, prompting SEO strategies to shift and adapt.

AI-Powered History Search

Google recently introduced an optional Chrome feature that leverages AI to help users search through their browser history and locate previously visited pages. This enhancement simplifies the process for users to revisit sites they’ve already accessed, potentially increasing return visits to those websites.

Chrome AI History Searches Through Page Content

Addy Osmani, a Chrome Engineering Leader, provided a detailed description of Chrome’s new AI History feature, revealing previously unmentioned details about how it operates. Unlike traditional browser history searches, which rely on URLs and page titles, this AI-powered feature analyzes both text and images on a webpage to help users rediscover previously visited sites. In an example shared by Osmani, the AI was able to locate a page based on its image content, demonstrating how the tool goes beyond conventional search methods.

He provided an example of a page he viewed that was connected to shopping:

What does he mean when he says that the word “England” emerged solely in an image? He did not specify whether the word was part of the image metadata, such as an alt tag. I initially assumed that “England” was included in the image’s metadata. So I found the identical page from the video in his LinkedIn post, analysed the source code, and discovered that “England” was not included in the metadata. If you watch the video, the AI Browser history shows many pages; therefore, it’s probable the AI simply ignored the word “England” and returned results with only partial matches. However, Osmani stated that the outcome was brought to light as a result of the image.

Here’s a screenshot from the video:

Here are the AI search results, which include numerous pages:

The figure above shows that the AI history returned more than one page, and the majority of those pages had nothing to do with an “England” shirt—only one did. This implies that the AI may have found the “England” website not because the word was in the image but because it matched similar phrases such as “Burberry” and “Sweater.” However, it is unclear whether “England” was part of the image data, which requires more explanation. Osmani also provides two other instances of how keywords from page content might be used to retrieve web pages that a user has already visited.

AI Browser Search Guide

Google provides a help page for this new feature, offering tips that also shed light on how the AI-powered browser search functions.

Key Takeaways

Chrome’s AI search promotes repeat visits through natural language queries. However, when users input simple text, Chrome defaults to basic keyword matching with the page title and URL.

• Exact keywords aren’t required.
• URLs aren’t necessary.
• Simple text queries are matched with the title tag and URL.
• Keywords in titles and URLs that reflect how users remember a site (its topic) remain important.
• The ability to rate search results indicates that this feature will continue to evolve.

Chrome AI History is a powerful tool that is likely to gain prominence as more people become aware of it and grow accustomed to AI integration in their browsers and devices. While this doesn’t suggest a need to overload metadata with keywords, it does highlight how SEO is evolving to adapt to AI’s increasing role in delivering web content.

Find the post on linkedin below:

https://www.linkedin.com/posts/addyosmani_ai-machinelearning-softwareengineering-activity-7236983842555764736-cd2R