The Future of Browsing Begins Now

It was only a matter of time before Google started turning its market-leading Chrome web browser into an “AI browser.” Today, Google began rolling out Gemini in Chrome, starting with Mac and Windows desktop users in the US. It comes with a bunch of new AI functionality — including AI Mode from the address/search bar, integrations with other Google apps (like Calendar, YouTube and Maps), and the ability to ask questions about the web page you’re on.

These features will also be available on both Android and iOS soon. In addition, Google is promising to add “agentic capabilities” to Chrome over the coming months, so that you can request an AI agent to take actions in the browser on your behalf.

Parisa Tabriz, GM of Chrome, remarked in a blog post that the new AI features are “about fundamentally changing the nature of browsing, and moving from a passive experience to a more proactive and intelligent one.”

Perhaps more concerning for web publishers, Tabriz added that it’s “about creating a browser that goes beyond rendering the web, to one that understands it, helps you be more productive, and keeps you safer online.” In other words, Google is turning its browser into a much more active middleman between web publishers and web consumers. Why “render” the open web when you can just talk to Google’s AI?

But as I mentioned, turning Chrome into an AI browser was inevitable. Google had already been adding AI into its dominant search product — first with AI Overviews (which summarizes web content), and then with the more interactive “AI Mode” feature (a full AI chatbot). The next step in the AI-ification of Google was always going to be its browser.

Google isn’t alone in building an AI browser. In late July, Microsoft launched “Copilot Mode” on its Edge browser, which is now being promoted as an “AI-powered browser.” Meanwhile, Mozilla has been baking AI into its Firefox browser over the past year or so, and AI-first companies like Perplexity and The Browser Company (recently acquired by Atlassian) have both released AI browsers. As for OpenAI, it has hired ex-Chrome engineers like Ben Goodger and is strongly rumored to be building a browser.

“We’re building an AI-centric browser that uses context — like the page you’re reading or the tabs you have open — to help you get things done faster, easier, and more safely than ever,” commented Chrome boss Tabriz.

One of the more intriguing evolutions to Chrome is to the address bar, which Google had re-named the “omnibox” in 2008 when it first launched Chrome. Up till the launch of Chrome in September 2008, the address bar was where you typed in a URL; for example, www.google.com. The omnibox made it possible to type a search query into that bar, too — a genius move, because it turned the most used input box in a browser into a doorway straight to Google’s search engine.

Well, now Google has added AI Mode, which it calls “our most powerful AI search,” to the omnibox: “You can use AI Mode to ask complex, multi-part questions, from the same place you already search and browse the web. You can then dive deeper by asking follow-up questions and exploring relevant web links.”

This is great for Google, as it makes it much easier for users to default to AI Mode. Again, not so great for publishers, as it throws up yet another roadblock to the open web.

The End of the Browser As We Know It?

One has to wonder if “browser” is even the right word for what products like Chrome and Edge are evolving into. We are moving further away from curiosity-driven exploration of the web — the modern browser is becoming an automaton, narrowing what we can discover and reducing the serendipity.

Let me briefly take you right back to the dawn of the web, when Tim Berners-Lee created the first web browser. His original browser application was actually a browser and editor, because it enabled users to both read and write webpages. Unfortunately, no one other than Berners-Lee programmed that “write” functionality into the first wave of browsers. By the time Mosaic became the first mainstream browser a few years later, browsers were “read-only”. Still, that was enough to encourage us to explore a world of creativity on the web, and also, Mosaic brought us multimedia. So, life was good.

When Chrome was released in September 2008, it made browsing both faster and safer — it was a significant step up in browser technology at the time. But the basic paradigm was the same: The browser was for exploring the open web (helped by Google’s search engine pointing you to the best links).

“The browser is no longer just a window to the web; it’s an intelligent partner that learns and adapts to your needs.”
– Parisa Tabriz, GM of Google Chrome

But now, with Chrome in 2025, that paradigm is changing. I’m worried we are losing the “read” part of browsers, too — or at least, the motivation to “read” webpages directly. This new Chrome wants to summarize the open web for you and do the browsing itself (which is what “agentic capabilities” really means, in practical terms).

Google Chrome GM Parisa Tabriz concludes her blog post with this observation: “The browser is no longer just a window to the web; it’s an intelligent partner that learns and adapts to your needs.”

I would go further. It feels like with Gemini in Chrome, Google is closing the curtains on that window to the web. Sure, you can still open those curtains if you want to — Google isn’t preventing you from doing that. But from now on, I suspect most of us will be content to stay indoors and let our agents surf the web for us. Why go outside, into the open web, when Google’s AI delivers everything you need?