The AI Browser Wars Have Begun
A smarter front end for the web is coming—and with it, a very different internet economy
I remember the first time I tried Firefox, and later Chrome. Each time, a faster, more user-friendly browser won me over and within days I’d permanently switched.
That kind of shift doesn’t happen often. But it’s about to happen again.
This time, it’s not about tabs or faster JavaScript engines. It’s about browsers powered by AI that not only load pages, but understand your intent, summarize content, and even act on your behalf.
And the ripples won’t end at the user experience. This transition will reshape the web economy: how traffic flows, how publishers make money, how ads are delivered, how marketers reach people, and how users get things done. It’s the start of a new browser war, with broader implications than the last.
From Netscape to Chrome: how browser battles were won
We haven’t seen a serious shake-up in the browser market in over a decade. Chrome launched in 2008, and by 2012 it overtook Internet Explorer as the dominant desktop browser. It did so mostly by being much faster and riding Google’s enormous distribution machine.
Every time someone searched on Google, they saw a Chrome prompt. Chrome was also bundled with antivirus software. And beyond marketing, it outperformed. It loaded pages quickly, updated silently, and gave developers better tools.
That combination—distribution, speed, and user experience—has driven every major shift in browser history:
Netscape rose because it brought the graphical web to life and made the internet feel exciting and new.
Internet Explorer won because Microsoft bundled it with Windows and made it the default.
Firefox won converts with tabs, customization, and faster performance.
Chrome crushed everyone with speed, simplicity, and massive reach through Google’s ecosystem.
And now, the cycle is repeating—only this time, with a new layer of intelligence woven into the browser itself.
The rise of AI-first browsers
Today, a new wave of AI-powered browsers is emerging: Perplexity Comet, Brave Leo, Browser Company’s Dia—and soon, almost certainly, one from OpenAI. These browsers promise not just incremental features, but a different relationship with the web itself.
They’re:
Faster: By blocking ads and trackers, and potentially preprocessing pages with tiny models to clean up layout, they could feel dramatically faster than Chrome.
Smarter: They can summarize, autofill, prefetch, and even act on your behalf.
More useful: Instead of just navigating, they help you get things done.
If OpenAI’s browser launches soon—as expected—it will land with an enormous advantage: access to over 800 million of ChatGPT users. Just like Google used its search page to push Chrome, OpenAI can promote its browser inside ChatGPT, where users already search and click.
That’s how the next browser war begins.
A shift with far-reaching effects
If only a small percentage of ChatGPT users switch to OpenAI’s browser, that’s already tens of millions of people—40 million if it’s 5%, which is already about one-fourth of Firefox’s user base. The impact of even that modest shift will be immediate:
Publishers will see traffic drop, as users summarize pages instead of read them, and have their chatbots answer questions based on website content instead of browse it. Publishers will also increasingly be serving pages to agents, not humans, as browsers take actions on users’ behalf. And they’ll lose data on all of this if AI-first browsers block tracking the way Brave currently does.
Advertisers will lose ad inventory as human website traffic declines. They’ll also become increasingly concerned about advertising to agents, and seek ways to prevent wasted inventory and accidental agent clicks. It will be even worse for them if AI-first browsers facilitate easy ad and tracker blocking. And this will happen at the same time search is becoming answer-driven, and traditional SEO is morphing into GEO—generative engine optimization—where visibility depends on surfacing in AI summaries rather than ranking in search results. The result is a double blow: falling organic traffic from both search and browsing, with fewer ways to target audiences effectively.
Users will get more done, with less effort, as their agents browse, read, click, and buy for them. They may also find themselves hooked on screens less, and able to assign their browser to do tasks while they go do something else.
The desktop web will change first. Apple has already opened iOS to alternative browser engines in the EU (iOS 17.4, March 2025). If that change expands beyond Europe over the next year or two, the mobile shift will accelerate. And OpenAI’s upcoming hardware—being developed with Jony Ive—could offer a clean-slate, post-smartphone experience that accelerates this transition.
Meanwhile, AI-first browsers will gather user data, learn preferences, and personalize results. They’ll offer users better performance and better answers. They may offer publishers new ways to plug in, perhaps through AI-optimized content feeds (I’m waiting for the first publisher to offer content only as Markdown or via MCP servers). And for advertisers, inventory could shift from websites, to AI-first browsers and their associated chatbots.
A new ecosystem is emerging.
What happens next
We’re on the cusp of a shift that feels profound but also still uncertain. Will Google lean into the trend, for example, and disrupt itself, even though it relies on advertising from publisher websites about to be disrupted?
We won’t have to wait long to find out. If reports are accurate, OpenAI’s browser could be weeks away.
I’ll be trying it the day it drops, just like I did with Firefox years ago.