The Mozilla Cycle, Part Iii: Mozilla Dies in Ignominy

It does not count as "easy" if the features don't stay disabled.

The problem with AI integrations in Firefox is not in whether they could be disabled or not.

Given that Mozilla Foundation isn't swimming in cash, "investing" in AI (a well known money sink) makes very little sense and will definitely undermine the development of their core product (the freaking browser).

Also, the timing of their Nov. 13 announcement is pretty bad. There is already chatter that AI may be a bubble bigger than the dotcom bubble. For a company that doesn't have deep pockets, it would be prudent to take the back seat on this.

It’s also trivially easy to disable ads in the Windows start menu, but the fact that they’re even there is shocking.

I use Firefox because I want to do at least something to keep the web browser market from becoming a monoculture again, but they’re making it increasingly hard to justify.

> It looks like till then google will be propping up mozilla to avoid looking like a browser monopoly

This is less and less the case each year. Historically, Google's accounted for over 95% of Mozilla's revenue. But through the recent launches of a bunch of products it's gradually knocked that number down to under 70% and seems to continue decreasing rapidly.

I often see two demands made of Mozilla: (1) focus on Firefox; (2) become financially independent from Google. IMO these two goals are going to be in conflict with each other. They started their own VPN, launched MDN Plus, etc in an effort to improve their financial independence. The AI gimmicks feel like they're in the same thread. I don't like it and don't ever wanna use it but I can't fault Mozilla for exploring that option.

Based on independent audits they are accomplishing (2) and based on their amazing performance in interop-25, interop-24, etc they are also accomplishing (1) as best they could.

It's pretty clear opportunists displaced the software ideologues at Mozilla a long time ago, but I still find the products to be more palatable than alternatives. It would take a long time to burn off all relevance of Firefox and Thunderbird even without adequate maintenance.

It is not about disabling AI; rather all the made effort for AI is away from something else.

> It looks like till then google will be propping up mozilla to avoid looking like a browser monopoly

The US government (via the courts) has sent a very clear message to Google that they can be as monopolistic as they want without consequences. I cannot imagine Google will continue supporting Mozilla for much longer.

> but also it's trivial to disable the AI stuff

That's the problem: the number of things you have to disable to stop Firefox--a browser that sold itself on protecting user's privacy--from spying on you, or facilitating the spying of third parties, keeps growing with successive versions, to the point that you now need a running checklist of what to about:config (thanks to the "simplified" Privacy & Security window) and extensions to add.

I still use it only because Chrome is worse, and even with Chromium, certain extensions either aren't available or are a bitch to use.

And what's increasingly annoying is that many web sites (for example, banking and utilities) don't even bother properly testing for Firefox support, given how obscure a browser it has become. Your traffic is also much more likely to be flagged as "suspicious" as a FF user, so expect to spend much more time playing "spot the traffic light/bicycle/bus/stairs" games.

I also liked Pocket integration.

But the naive purists seem determined to team up with the genuinely evil in every walk of life, so Chrome monoculture seems inevitable.

And it's not like Google or Microsoft is going to do anything with AI that is worse than this, right?

Same here. I find it useful, and it doesn't feel like it took several engineering years to build, diluting Mozilla's focus from "just building a browser".

Actually, whenever I hear people argue that that's all Mozilla should ever be doing, I wonder if they really mean a HTML and Javascript engine? While that's important, browsers are more than that; the chrome matters too.

Same, really don't understand what all the hoopla is about. AI integration in Firefox is inevitable, as is with all other browsers.

You can already do what you're looking for by reading the browser cache as new data is cached. This would allow you to see the site as it was loaded originally, instead of simply fetching an updated view from a URL. The data layout for the cache in Firefox and Chrome is available online.

They'd probably reject that idea under some bullshit privacy or security excuse Wayland-like reasoning. Also why we don't have XUL extensions anymore and why they'll eventually copy chrome on that manifest crap.

My feelings are the same. Firefox is barely alive, and what little sustains it is goodwill and people not wanting to deal with Google more than is needed. Yet they keep on eating into their goodwill and do seemingly nothing to build it back up again (unless introducing AI features is somehow supposed to be that). Every bad feature implemented could have been a bug that got fixed instead, or effort to push back against Google. And yet they consistently opt to put their effort into features that push people away from them, and don't put that effort into things that would at least retain the people they already have. It's a lose-lose situation.

Agreed. I just don’t want to have to use a Chromium-based browser given Google’s grip on the project - but what alternatives are there if Mozilla goes under?

Safari? Orion?

I genuinely do hate this too, I have to reset it every time. But it's worth 600 million to them so I'm kind of begrudgingly bearing it.

> But in the real world, the Web is a pile of auto-generated and auto-assembled fragments of slop

There are parts of the web like that but your assertion seems to rely on this being universally true. It clearly and obviously isn't.

> Also in the real world, people are decisively choosing the AI-generated summaries and fevered imaginings.

Are they "decisively" choosing it if it's turned on by default? If it were actually opt-in then we could measure this. As it is I don't think you have any data to rely on when making this assertion.

> Not for everything, but web search -> URL -> page visit is becoming a declining percentage

The same web search companies that own AI models they're trying to sell? Do you not suspect there could be a few confounding variables in this analysis?

> except that automatic translation is really nice

Which we already had and has nothing to do with language models masquerading as "AI".

> is fine and dandy but ultimately pointless unless you have an alternative that doesn't require the entire world to cooperate in turning back the clock.

An alternative to what? Tab renaming? Bad article summaries? Weak search engine algorithms?

It's certainly wrong on some things. For example, claiming that there are no examples of positive uses of LLMs is flat out wrong. I'm generally on the very conservative side when it comes to AI hype but even I'll admit it will be transformative in some ways.

Certainly not the insane fantasies most of the gen-AI CEOs are pushing, but, for example, it's clear beyond any shadow of a doubt that traditional search is dead. AI supported search is far superior in every way. For clarity, I'm talking about "deep research" style search where you get verifiable links to source materials along with your answers.

It's absolutely not crazy for Mozilla to get into this space, though I think if I were them I'd make an adjacent "AI search and agnentic AI first" web application adjacent to Firefox and keep Firefox as a legacy style browser. This would both give Mozilla a clean slate to do it right, while also keeping happy those who are not early adopters.

I think the vast majority of attempts to shoehorn ai into the browser are deeply unimaginative and very much garnering a "who ordered that?" reaction. But I suspect we're going to converge on some specific use cases that everyone's going to want and it might just be important to be in the game now as we collectively figure out what they are.

Recently some Ycombinator funded project got highly upvoted on HN, a Chrome based extension that used LLM capabilities to effectively do grease monkey style scripting live in response to human requests. Now that is interesting, and it's a specific application that's actually meaningful and it's not just another AI chat sidebar.

I think it's a matter of workshopping but I bet we're going to be discovering things users actually want that are not yet obvious to us. The example I keep thinking of is non-stupid agent tasking. I wouldn't mind an agent that browsed Amazon for Kindle unlimited hard sci-fi books with critical acclaim. I would be willing to be there's going to be numerous "whybdidn't I think of that" uses cooked up in the next few years.

> the continued existence of at least ONE other browser.

thankfully next generation of browsers are here - ladybug and servo, so at least something will survive even in the worst of the worst cases

I will stick with Firefox due to multi account containers. Chrome does not offer a comparable alternative, and this extension makes working with AWS significantly easier.

https://addons.mozilla.org/en-US/firefox/addon/aws-sso-conta...

Pretty sure i read that the translation is not AI (LLM) based

> the automatic translation support is very helpful.

It's great, and I love that it works fully on-device, i.e., it's as privacy-preserving as it gets.

Yeah I think although he acknowledges how biased the Firefox forum poster sample is, I think he is vastly underestimating just how biased.

If Firefox is reduced to just nerds who post in forums it's totally dead. Maybe it's at that point already.

Tbh I still use Chrome because Firefox's performance just isn't as good. I wish it weren't so but there we go.

What useful, non-creepy AI features are there?

I cannot imagine why one would want AI in a browser at all, but I am open to the possibility that there are applications I have not considered.

What features? Voice control?

If it is about chat, do we actually need Firefox adapted when you can go to gemini.google.com or some other one and write what you want there? Optionality is ensured since you actively have to go there.

A relatively small variation for a client-side LLM would be something like: "suggest variations on this search that might produce better results"

Had this issue the other day that I wanted to try to find a remembered example of using Ground Penetrating Radar to find hidden roads in Northern Spain and Southern France, yet for everything tried I could not get Google to actually return results that would ever focus on what it was I actually wanted.

There was some GPR results, there was some about hidden structures, yet finding the actual scholarly articles discussing hidden roadways and searching under farm fields for features that reveal or point to their existence simply never returned positive. Rephrasing, changing the sentence, trying to prioritize search terms never really changed much.

Doing stuff too like, "image search, pixel tree, except filter out all the pay for an image stock photo websites." Really difficult to find anything these days cause there's so much chaff of stock image websites.

Google's AI overview regularly hallucinates or gets the answer wrong. Obviously if you're going to run inference on every search from billions of users, it has to be a very cheap model.

Half the time this is just because of Google websites having poor performance.

Microsoft at least has the decency not to base their entire marketing on being pro-online-privacy.

> demonstrated by their ever-shrinking browser market share

At this point I think Firefox market share stopped being in Mozilla's hand.

Just as it was during the browser war days, the critical issue went back to site compatibility: Firefox performs poorly on Google properties (Gmail is fine, YouTube, gsuite, admin consoles are pretty bad), and document based services like Notion or Figma. It kinda works, but Chrome based browser perform notably better.

The main point of course is that those sites are at fault (sometimea intentionally when it comes to Google), but that doesn't change Mozilla's position. Stop using Google services is just not a great choice, and many of us use them rely heavily on them for work.

Mozilla could make technical miracles and or bring some incredible feature from the left field, but that's a tall order for any company that size, so I'd expect most of their future effort to still end up with lower market share, whether or not they had good ideas.

  > Do you think Chrome gives a shit about the Internet? No.

  >> This has the unintentional side-effect of strong-arming the W3C into compliance with the engine and not the other way around.

    Firefox performs poorly on Google properties 
    (Gmail is fine, YouTube, gsuite, admin consoles 
    are pretty bad)

But they already do?[0]

--

[0]: https://cyberwarzone.com/2025/11/07/mozilla-unveils-enterpri...

As others have mentioned but not yet shared links, enterprise support is coming in January: https://support.mozilla.org/en-US/kb/firefox-support-organiz...

100% agree. I would happily run a dedicated enterprise browser that blocks downloads, has DLP, has watermarking (etc) if it meant I could use my own PC. Not Browser Isolation or VDI - An actual enterprise browser.

My job is pretty much 100% in browser though, so I realise this isn’t viable for everyone.

The main driver of market share loss was the rise of mobile and Chrome being the bundled default on over a billion devices. I don't think Mozilla's run has been perfect, but I am flabbergasted at so many confidently wrong retellings of the history of market share change that treat it exclusively as a story of Mozilla's strategic missteps and make no space for the fact that a trillion dollar company with the #1 or #2 most visited site in the world for the past 20 years muscled a new browser into the picture, leveraging their monopoly in search and mobile, plus laptops produced and sold practically at cost. Firefox could have made every perfect decision you could have dreamed of and still suffered a market share collapse.

In fact, I think there's a pretty clear-cut example of a natural experiment demonstrating what it looks like to execute in the browser space to nearly perfection and still lose. In my personal opinion, Opera at its peak with the Presto engine, represented the most impressive combination for it's time of elite level performance and stability, genuinely good innovation that benefited the user, and commitment to the core browser above all else. My favorite was whenever they rolled out Opera Unite (to this day a truly mind blowing idea), I think Opera 11 or 12.

And, at a time when it truly mattered, was light on resources and bandwidth and even shipped a portable executable that could be run from any Windows PC from a USB stick. Not only that, but it was consistently ahead of the competition with embedded and device adapted versions. Is business partnerships were creative and cutting edge too. They were early to mobile, struck deals with oems, and even got on the Nintendo Wii. They offered paid and subscription options. IIRC they sold "speed dial" placement for ads and got into the search licensing game. So in everything from performance to speed to stability to innovation that actually benefited users, to intelligent business partnerships and experimenting to find revenue, Opera executed at perhaps the best level anyone could, the perfect moneyball browser. And I was never originally an Opera fanboy, I preferred Firefox 1 and 2 at the start, but pivoted to Opera because, as a college kid with no money, it delivered an impressively modern experience on lightweight hardware.

Despite executing at the highest level both in software and business decision making, it didn't matter, because distribution power trumps product quality. Sustaining a fully independent rendering engine became financially unsustainable, and with the maturity of Android, carriers favored bundled stock browsers.

With no options left Opera then made what I consider a difficult and very unfortunate decision, but perhaps having no other choice, sold to a new ownership team, pivoted to Chromium, and lost much of its team to Vivaldi which is also based on Chromium. But at no point in the story was their loss of financial visibility or market share due the loss of vision that people think explains Firefox's loss of market share. If the world actually worked that way we'd all be using Opera 25 right now.

Edit: If someone more knowledgeable could chime in, I would be fascinated to know if choosing a browser on Android could be a potential monopoly remedy. There's already precedent for that on Windows, iOS and on Android in EU for search.

They actually have an Enterprise Edition [1]

[1] https://www.firefox.com/en-US/browsers/enterprise/

Features: [2]

  - Rapid Release or Extended Support Release channels
  - MSI deployment wrapper
  - ADM and AMDX group policy templates
  - MacOS plist file policy templates
  - Linux target JSON file “policies.json” policy templates
  - “Open in IE” extension for ActiveX

Does not have the built-in DLP you're requesting (at as far as I could find) and Firefox already has pretty aggressive adblocking that sets off lots of sites for me.

[2] https://support.mozilla.org/en-US/kb/firefox-enterprise-vs-n...

>I wish Mozilla would explore the enterprise productivity space.

They're rolling one out in January 2026

They don't wanna be Redhat - but IMO it's their biggest mistake (I worked there 10 years).

Palo Alto Networks is one of several companies pushing custom versions of Chrome as enterprise security browsers doing exactly what you're talking about: holistic DLP/anomaly detection, URL filtering and content inspection in-browser. Presumably because it's closer to the malicious behavior, and network MITM is harder to accomplish with newer TLS and with decentralized workforces.

If Firefox had a more customizable, enterprise-feature-focused browser maybe we'd be seeing it used instead of Chrome? I don't know.

Time-to-market favors chromium. It's also by far the most commonly used engine for headless browsing, creating lots of useful prior art.

> But Mozilla had a good run.

I don’t think Mozilla is over.

I also don’t think people should equate their history with their current state. They lied to their users and told them they’d never sell their data, and then they did. That is much worse than never having made the promise. I don’t trust them.

But, they have far too much support and are far too embedded to disappear anytime soon.

As the downcycle lingers on everyone’s doorsteps a lot of non profits that do incredible work are losing funding as the investors back out from commitments they confidently put forward when things were going well. It quickly shutters the non profits. It makes me sad tbh.

Can’t say this is the same thing that happens at Mozilla but you are very right in that a lot of nonprofits seem to be lead by those bring the same operational decision making experience and solutions that you see in publicly traded companies. There are plenty of non-profits that are indistinguishable from public companies in how the board is composed of an inner circle of wealthy unsavory people.

Mozilla management have nobody to respond to.

For-profit (public) company at least have shareholders. Mozilla have zero motivation to improve aside from being retirement home for failing managers