But to your point, I think the simple reality is that LLMs are increasingly taking the place of search and so having all your funding based on search licensing might be risky when it's at least possible that we're going to be in a new paradigm sooner than later.
I honestly think AI in the browser right now is generally very half-baked and doesn't have any well thought out applications, and raises all kinds of trust issues. I can think of good applications (eg browse the Kindle unlimited store for critically acclaimed hard sci-fi books), but there might be better ones that I'm not thinking of. It just might make sense to be involved so you went caught flat-footed by some new application that quickly progresses into something people expect. And of course because HN commenters are famously self-contradictory in response to literally everything Mozilla does, it's a damned if they do damned if they don't situation: if they load AI into the browser it's pointless feature bloat. If they don't then they were sitting on their thumbs while the world moved on when they should have been reinventing themselves and finding new paths to revenue.
Unfortunately, we live in a time when anti-trust regulations mean nothing.
The fact that it's difficult to separate Chrome from Android dooms most competitors, which is bad for everyone.
Chrome even has significant user share on Mac OS; the numbers I'm finding are around 40%.
It's hard to guess whether people are much less inclined to switch browsers on mobile than on desktop, or if they just like Chrome. Either way, the odds are against anyone who tries to compete with it.
This is primarily because most people on Windows use Gmail and other Google services, and any time you visit a Google web property from a non-Chrome browser, there’s a prominent “Install Chrome” button that’s placed on those. Without Google’s web properties pushing Chrome even to this day, Chrome may not continue to be as big.
That lack of connection to tech giants is a strength in the trust angle. And I think they’re right to be thinking about AI: people are using it and there does need to be an alternative to tech giants/VC funded monsters
Will they be successful? The odds are stacked against them. But if they’re not going to even try then what purpose will they serve any more?