Firefox Will Ship with an "AI Kill Switch" to Completely Disable all AI Features - 9to5Linux
After the controversial news shared earlier this week by Mozilla’s new CEO that Firefox will evolve into “a modern AI browser,” the company now revealed it is working on an AI kill switch for the open-source web browser.
On Tuesday, Anthony Enzor-DeMeo was named the new CEO of Mozilla Corporation, the company behind the beloved Firefox web browser used by almost all GNU/Linux distributions as the default browser.
In his message as new CEO, Anthony Enzor-DeMeo stated that Firefox will grow from a browser into a broader ecosystem of trusted software while remaining the company’s anchor, and that Firefox will evolve into a modern AI browser and support a portfolio of new and trusted software additions.
What was not made clear is that Firefox will also ship with an AI kill switch that will let users completely disable all the AI features that are included in Firefox. Mozilla shared this important update earlier today to make it clear to everyone that Firefox will still be a trusted web browser.
https://9to5linux.com/firefox-will-ship-with-an-ai-kill-switch-to-completely-disable-all-ai-featuresOpen linkView original on lemmy.world
Why not just ship it without any of the AI stuff and give users the option to install and use it instead of bloating the application? This also confirms that the stuff is essentially OPT OUT instead of OPT IN
The bubble is AI and they want some of that bubble investor money is my guess, so they put optional AI
"On by default unless you run down a setting buried in a menu" is the thinnest type of optional in computing.
That's fair, but also if you search AI in the settings it shows you all the options
It’s like saying cosmic is optional on Pop_OS.
Sure, you can rip it out if you really try.. but is it really optional?
Have you attempted to turn of AI in Firefox? It's literally like two checkboxes and it's off
Flip the script man.
Imagine if enabling AI was like two check boxes and it's on, for those people who really want it.
Sounds great.
I mean yeah that would be better, I'm not disagreeing
Have you attempted to turn of AI in Firefox? It's literally like two checkboxes and it's off
In their defense a very tiny percentage of users even open options and of those an even smaller actually change stuff.
Maybe slighlty different for Firefox as probably more power user use it than other random programs. But basically if something is not enabled by default, it doesn't exist.
Because they're counting on people who know nothing about technology using the AI stuff when it's placed in front of them.
And also ..... will the kill switch turn off the AI entirely ... or partially? Since the AI system is baked in, will elements of it still operate in the background even if you turn off the switch?
Not sure what you mean by "will it operate in the background"? The current (and planned) features collect no data. They "operate" when you use them. Disabling them will remove them from the UI.
lol .... so they won't change how they function ... just remove them from sight
out of sight, out of mind, right?
Whenever I trust big corporations ... or even big organizations with a lot of power in their hands ... it's never usually good for common people like me and you.
No, they won't be used at all, and will be hidden from view, if turned off.
It's an open source browser.
The publicly available code is the most verifiable system of trust you'll find.
What he wrote doesn't seem ambiguous on this at all. But we'll see.
So you agree that it will be baked in and impossible to actually turn off. Yep.
Otherwise, they would have made it an extension, right? If it's optional, it needs to actually be optional ... that's what am extension is. That's the whole point of them.
You can not push the button that says AI.
You can also hit the kill switch that completely removes that button.
That's opt-in enough.
If it starts reading pages or doing things without you pushing a button, that's an issue.
And therein lies the rub. The question is whether or not people trust that it won't be doing that regardless of whether or not you hit the kill switch.
Good thing it's open source and we'll immediately see that they aren't doing the thing you're claiming.
No, you don't have to trust anything. It's open source, you can read the code.
And if you're feeling paranoid, you can compile it yourself.
No
Sounds like they will be opt in, not opt out
No, go deeper into that mastodon thread.
The dev has a really hinky defention of "opt-in" thats basically "yes we push all this on by default and realize it will be the norm for most of our users because of that, but you technically dont have to interact with it so thats opt-in."
Somehow, eventually having a buried menu option that "opts out" of AI is also part of how it will be opt-in as well? Its a self serving mess of rationaliztions and doublethink, no matter the claim on the tin.
I mean yeah, that's a fair point, and the dev said that themselves, that the definition of opt in is ambiguous. The definition they seem to use is that AI won't run unless you explicitly tell it to, and I think that's ok. There'll be a button that you can press to do some AI action and you can hide it using the kill switch.
I do hope the kill switch isn't hidden behind 5 layers of menus
Thats not ambuguity. AI will be opt out in firefox, which is them abandoning core principles like user choice and privacy.
They can do that, but playing like they aren't by redefining well established terms in UI/UX is disengenious, and cuts right through the "we will earn your trust back" messaging made by the same dev.
I think it's quite clear there's ambiguity (hence this discussion). How would you define opt in? Should a user not even see the button for an opt in feature?
I think the big defining question is what will the AI features that they will implement do exactly and how will they run. If it's something that runs in the background (even as unintrusive as the summaries on a search engine like DDG), then it's opt out by default as it's constantly running whether you want it to or not. If it specifically and exclusively runs when you hit the button to activate it and doesn't run at any other time, then I'd say it's unequivocally opt in. And regardless of what a company says that their software will do, at this point I won't believe it until somebody has done a full teardown and discerned what exactly it does behind the scenes. I've seen enough nonsense like the Epic Games Store accessing your browser history and recording keyboard inputs or whatever the other absurd incident was.
Nah, I think it should be optional. Some AI features may even be useful — like an AI script to get rid of AI slop or something, idk.
In my opinion there is no ambiguity at all.
Opt-in means that the feature is disabled by default and until the user enables it. This is NOT what Firefox will be doing.
Opt-out means that the feature is enabled by default and can be disabled by the user. This is what Firefox will be doing.
Whether the user actually uses or not the feature is not a factor in determining if it is opt-in or opt-out.
So if you never press the AI button, it's never enabled. It is opt-in in the strictest semantic sense.
What you say here applies for things that run automatically, like the anonymous usage reports, which is opt out, not for things you activate yourself.
A feature that will not do anything unless you explicitly press a button to start using it is quite literally opt-in, though? Opt-in doesn't mean "I won't even know the feature exists without hunting through the settings". It just means that it won't start doing things without your consent. Presenting a way to provide that consent in a more visible place than buried deeply in the settings does not make it opt-out. It might be a bit annoying to you, but it has no effect on your user choice or privacy, especially if there's also a way to globally hide it and any other features like it, including new ones that might be added in the future.
Let's have a look at how it works now, so we don't need to speculate.
When I configured Firefox for AI, I got to choose my LLM of choice. I chose Claude. Now, if I select some text, I get a context menu option that says "Ask Anthropic Claude", which branches into these options:
Notice the last one? That's not a "buried" option. That's as front and center as the options to use it. Mind you, if I decide to not use it, then nothing happens. The only thing that's changed is that I now have an optional shortcut for LLM features that open in a sidebar instead of a new tab.
Oh, the humanity.
I don't see why there is a big outrage. Sure I'm not a fan of the AI features and I certainly will disable them but it's tot like they're forced upon me. Some people like (want) AI in the browser and good for them, this makes the browser better and easier to use for them. For me, it doesn't change my experience at all
(Commented this separately on purpose)
Come to think of it, I do enjoy the translation feature in Firefox
The alt-text tagging is pretty amazing according to my sister (blind), too.
I’ve been thinking the same thing. The online tech community is a very small part of a much larger pie and they need to serve multiple audiences. As long as it can be turned off and truly be off, who cares?
People don't trust that it can be truly turned off and that it won't act maliciously in some way. That's really the crux of the whole saga. We're at a point where phone companies are getting survey results that say that 80% of users either don't care about AI nor use it or find that it actively makes their user experience worse.
Did those people forget this is am open source browser and they can actually check it's doing what it says it's doing?
And if they're that paranoid that they don't trust the pre-compiled binaries, they can just compile them themselves.
This discussion is completely absurd to me.
Quite honestly, I don't think the average person even knows what open source means. They just know that Mozilla, like every other company, is shoving AI into their product, and that AI has either been useless or actively harmful to their user experience.
This should just have been an extension. Having this as a core integration makes the browser have more surface area for attack.
If compromised, it won't be an easy fix like disabling/removing an extension.
Looks like execs behind closed doors are just trying to water down the Firefox brand until it's hollow and then jump ship.
Many people love AI, I have a lot of acquaintances who actively seek out the best "AI browser" whatever that means. It makes sense for mozilla not to fall out this bandwagon just yet.
Is there nobody with sanity left? This has blown up so much the user base clearly does not want it. Focus your efforts elsewhere. You gain marketshare by putting users first. Also fuck markets.
If all Firefox users donated to Mozilla it could work. Alas, we don't.
I probably would, if the organizational structure and its spending focus(es) weren't so fucked up. They have been spending insane amounts of money on bullshit like AI instead of core browser features, and their leadership has extremely high wages for something that should be a non-profit open source organization. And it has been like this for years at this point.
Their CEO makes more than I think CEOs should earn in general, but the rest of their executives earn relatively normal to low salaries for their roles and the sector.
Non-profit doesn't mean everyone works for free.
Well to be fair they only had that or onlyfans to get paid to sit there playing with themselves.
They'd have done less damage going for the OF option.
I mean, I bet they'd make a killing off of Firefox themed thigh highs...
You can donate to Mozilla, you cannot donate to Firefox.
Which is probably a good thing. I appreciate projects like Thunderbird as well.
No money you donate to Mozilla Foundation goes to either Firefox (MozCorp) or Thunderbird (MZLA).
They are separate entities.
Pretty much this.
They are making market based decisions because they have to, and all the users bitching and moaning about them making financially driven decisions don't donate anyways.
Then let's
How? Where? I'll donate, take my money, and ads a voting system where paying users can vote for the next features
https://www.mozillafoundation.org/en/donate/ if you're actually interested in donating.
I would donate if I ever had spare money but I never do.
$5?
Make fun of the poors yeah?
No, I just don't believe they really can't spare $5.
Maybe they can't but I'm skeptical.
They? You mean me? Lmao you are replying to the same person.
A very vocal portion of the user base, but we don't actually know what absolute portion cares. I'm personally unlikely to use possible AI features outside translation, but Mozilla has generally done enough that I don't feel particularly worried they're going to mess with my privacy or force me to use a feature I don't want.
They could do a survey amongst Firefox users about what they want.
But if the result is anti-AI they can't claim anymore that they weren't aware of their users opinions.
The issue is that there aren't many of us Firefox users left, so asking us while FF wants to get NEW users to expand the market share (which is badly needed, so they do not lose their seat at the table regarding web standards, and to make them less dependent on googles payments) is not helpful at all.
As long as i can switch it off with one click, i couldn't care less and will continue using FF, but as you can see many existing users will bitch and moan even if it's just one click.
This is a balancing act and Mozilla behaves like an elefant in a porcelain shop right now. Worst case they loose their current users without attracting new ones.
I'm one of them. Why not make it one click for people who want it instead?
And where to?
Ladybird, Servo and Floorp are all not useable as a daily driver and will take years to get there (and btw, the ladybird guy is a major shithead and last i heard of Servo was that they were going to cater to the embedded market, not a full blown browser).
Firefox forks can do what they want, even switch off the AI button, but i'd still say they help keeping the browser engine itself afloat, because they still depend on Firefox - there's not one fork with enough dev staff to keep up. That leaves us with chromium based browsers and safari. I'd say the commitment to the current userbase to make the changes optional is good enough to keep most of them.
I'd put current Firefox users much more in the department of "able to find the settings" than the vast majority of users. The majority wants something that works with everything they throw at it out of the box without rummaging through settings.
If both noteworthy browser engines are made by companies who make decisions against their user's interest I might as well switch to the one with higher development budget.
And where does AI come into play here? It's not like a browser without AI doesn't work.
At least Firefox isn't an extension of the worlds largest ad company, no amount of dev budget can fix that.
Context aware search, summarizing in side view or importing an agent directly from a repository into your browser are things that come to mind without much thinking, and i am not a developer.
Chill out. It's literally just a sidebar for your LLM of choice.
Don't like it? Don't use it.
Don't want it to clutter up your context menu? The same menu contains the option to disable it. Boom! Problem solved.
Gonna use Chromium-based with no µBlock because your feelings got hurt? Have fun.
We need valve but for browsers.
Funny. That's exactly what Mozilla used to be.
Yup, RIP
Floorp and Ladybird seem to be the two browsers people are looking towards to shakeup the browser market once they release.
Ladybird I follow since it's an entirely new browser engine and can help restore a little democracy to the web, but why Floorp? I'm looking through its website and it seems to be a more customizable Firefox, which is nice, but doesn't seem particularly revolutionary (and forks of Chromium/Firefox are kinda a dime a dozen).
No idea why Floorp has garnered attention but other seem interested.
Ehh, I'd pass on Ladybird. I've been donating to Servo myself.
It seems to me that the issue is how we consume the web versus games: We're used to pay to play but not to browse the internet. Valve is able to make money without relying on affiliations or donations.
Getting stuck on Lemmy can get us into an echo chamber. A lot of the mass public actively uses AI and may even appreciate these features.
[citation needed]
This sounds like an opinion from the LinkedIn echo chamber.
According to top results for a few searches,
ChatGPT has 700 million weekly active users, as of September 2025.
Gemini has 350 million monthly active users, as of April 2025.
ChatGPT is the number 1 app on the Google Play top downloaded apps currently.
Some numbers may be slightly inflated for a number of reasons, but another source is speaking to people IRL. I often overhear conversations mentioning ChatGPT or AI in general.
"AI bros" are definitely cancerous but a lot of average consumers do in fact use AI frequently
Does anyone even talk about what the “AI features” are?
Could I, liked recolor webpages? Automate ublock filters? Detect SEO/AI slop? Create a price/feature table out of a shopping page?
See, this would all be neat like auto translate is neat.
But I’m not really interested in the 7 millionth barebones chatbot UI. I’m not interested in loading a whole freaking LLM to auto name my tabs, or in some cutsie auto navigation agent experiment that still only works like 20% of the time with a 600B LLM, or a shopping chatbot that doesn’t do anything like Amazon/Perplexity.
That’s the weird thing about all this. I’m not against neat features, but “AI!” is not a feature, and everyone is right to assume it will be some spam because that’s what 99% of everything AI is. But it’s like every CEO on Earth has caught the same virus and think a product with “AI” in the name is like a holy grail, regardless of functionality.
Right right. If they had real innovation, they would have defined it clearly as you suggested. But they didn't, so they don't. It's all snake oil, again, because that's the entire AI industry.
The term snake oil is actually especially fitting for this, due to its origins.
In Britain in the 1700s there was a somewhat common recommendation for using rattlesnake oil from the fat of the snake for skin diseases/rheumatism. The efficacy is debated but it's got some amount of potential for change (if not help).
This turned into people in the US selling mineral oil as "snake oil" as a total panacea. So a product that actually could do stuff being used as the poster child for a completely useless product that can solve every issue ever, buy as much as you can today.
Snake oil indeed.
You reminded me that one use for AI I'd really like is removing all photos of Trump, Musk and Putin from my screen. Another is filtering the twenty reposts of every event in US politics and the incessant whining about prices. Alas, I need these in phone apps more than the browser.
You don’t need LLMs for that. An iPhone is plenty powerful enough for image recognition and text classification.
That’s sorta the funny thing about AI. There’s tons of potential, but it’s just unimplemented. Even on PC, you pretty much have to have some Nvidia GPU and fight pip setting up python repos to get anything working.
That stuff is commonly included in the AI umbrella.
Eh, I draw a distinction between oldschool visual recognition and matching some keywords, versus full-blown LLMs. I used 'AI' to mean the latter in my comment above, as intended by the post itself. I also have doubts about the effectiveness of the older approaches in regard to the uses that I mentioned.
I’d like a AI feature that bounces back on any ads or intrusive crap including propaganda. But AI is being pushed by the same people so if it happens it won’t be genuine and it will just evolve to give the illusion of not being pushed.
FF, any browser, any social media platform , I can select ‘I don’t wanna see this’ or go adjust settings to disable but nah, an update later and it’s back, or it comes back under another form.
It’s still manipulation. And I can’t trust it to manipulate ever.
you don't even need AI for that and can do it on your phone. Piefeed for example asks you when initially setting up an account "Hey do you want to see stuff about Trump, Musk, etc? no? cool we won't show you that" and that's it. works great.
What indicates that Piefed does this on the client side?
I would use it to block all the people whining about AI
So, nothing.
Yeah, basically a nothingburger.
Can’t wait to power browse.
Are we goong to get power browse or drunk browse?
The last feature is the mildly interesting one, but in my experience just not useful enough to do much, even on specific browsing finetunes or augmented APIs.
I guess shake to summarize is mildly interesting, but not really? I simply can’t trust it. And I can just paste the (much more concise) relevant text into a chat window and get a much better answer.
I agree. They're quite vague about what the feature would actually do, and I have a hard time believing that I would use it.
I think it would probably be wiser spending time and effort on other things (like a really good built-in Dark mode or better memory management), but I don't fault them for experimenting. Worst case scenario they make something that sucks and either remove it later, or you can fork it off.
... I feel I have an idea of what this means, but it still breaks my brain just a little bit.
The one I use the most is their offline translation. I don't have to send my data to Google Translate.
My sister (blind) uses the new screen reader stuff a lot.
Mozilla is certainly adding good AI features, but the chatbot integration isn't something I have much use for.
Translation and screen reader have been a solved thing for a while, no “AI” browser necessary. I’m all for nice features, but bolting in a chat bot that phones home with activity data ain’t one of them.
This is an excellent point: there are potential features I wouldn't mind trying out. But of course those features aren't available, because aren't the features that Mozilla leadership's buddies in tech are pushing, and often work against what big tech wants.
The reason the "kill-switch" wasn't made clear originally was because it literally didn't exist until users very vocally tool them where to shove their AI crap.
It was added on afterwards.
What? They've been talking about features that are now being called the "kill switch" for the better part of a year. Literally all they did that's new was give it a dumb name.
Just to point out that per the discussion in the screenshot: Synthetic datasets are typically generated from models that were trained by poverty-pay Kenyans. This is basically ethics-washing.
That's literally not true, though? They've spoke about it for ages.
The real issue is not whether we are going to be force-fed this features or not, but the fact that a foundation with limited resources is going to spend any sizable amount of them developing a solution its users are not interested in.
Waiting for Ladybird at this point.
Librewolf
These guys played host to one of the tech world’s most prominent homophobes for decades. I like the browser but the foundation has been trying to fuck everything up since the beginning of the web.
Huh? The board ousted him when his political donations came out
According to your link below he was co-founder of Mozilla in 1998. Based on the other information on that page, he had a very significant role in shaping Mozilla and their tools, so as disagreeable as his personal views may be, it's not impossible that Firefox might not even exist today were it not for his work there.
Someone else has already pointed out that he was pushed out, but he actually resigned due to public pressure (he was only CEO for 11 days, and one of the board members even left due to him being appointed) before going on to found Brave and becoming the CEO there lol.
If I chose not to use products based on the personal beliefs of the people who worked on them I don't think I'd have very many options. Mozilla has made heaps of questionable decisions over the years, but the other options are generally much worse.
Obviously, as a Firefox user myself, I’m making similar choices.
But I’ll be dammed if I let an opportunity go to waste to remind people that Brandan Eich is disgusting human trash. He’s responsible for JavaScript, and that’s only least of his sins.
Yep I got the feeling that we were of similar minds about it but I wanted to provide some balance in case anyone else was weighing up switching to a different browser.
I knew JS was his handiwork, but I wasn't aware of his significant role at Mozilla or that he is also behind Brave, so thanks for sharing that link.
Out of the loop, who was that?
https://en.wikipedia.org/wiki/Brendan_Eich
Eeesh, and an anti-masker too 😟
Repeat after me. "There is no such thing as a non physical kill switch"
I have a better kill switch: Waterfox and LibreWolf. Don't have to worry about of that nonsense right out the gate.
👍
Zen has also committed to not include AI features
I jumped to LibreWolf this week. Really like it, it looks acat and feels the same. But I trust it more. Been a FF user for over 10 years.
I'm considering it. The only reason being to get away from a corporate stance that could shift at any time, even though I don't think it's quite there yet.
However on issue of Firefox going the way of all the other browsers, I swear that the last update or so of Firefox asked me if I wanted to enable AI, I said no, and it told me how to turn it on if I ever wanted it. Much like when I first used DuckDuckGo. So wasn't that opt in? Did it change how it prompts a new user?
It’s worth a try. It’s only a few min to do. My extensions seem to be all from Mozilla and working fine.
The only real difference is it has a bit of a classic FF apathetic and seems to highlight who’s abusing privacy. I can click that off but I like it.
I also use DDG but that’s gone downhill on its results in the last few years.
Some websites doesn't work with librewolf like Onshape.com . Is there an easy solution for website like those?
Waterfox. As I understand it, Librewolf goes to extremes for privacy to the point where it may impact your browsing experience. Waterfox cares about privacy too but not to the same extreme.
But that's just saying that instead of using Firefox and not turning on the feature, you'll use a less maintained version of Firefox where they didn't enable the feature. I don't feel like those projects have much value add in the privacy spectrum compared to Firefox, particularly when one of them was owned by an advertising company, and neither of them actually has the resources to maintain or operate a browser in isolation, which is a major concern regarding security and privacy both.
While I can't speak for LibreWolf, I can tell you that Waterfox is based on the latest ESR builds and is extremely well maintained to the point of evolving into its own thing entirely. It's one of the oldest forks I've known. The fact it's been around this long should speak volumes. That being said, most modern forks that I've tried tend to base themselves on ESR as well and evolved in a similar way.
Well they're clearly not taking it all that seriously as it should be an Opt-IN feature, not an Opt-Out. They're banking on a majority non tech savvy userbase to not even bother disabling it. fine, whatever, that's on the user.
But it's just more Firefox bloat that I have zero desire to deal with. If I wanted bloat in my browser I'd go use Vivaldi.
Not buying it. Kill switch will migrate further and further into about:config until it eventually too goes away without notice in an update six months from now.
No that's way too paranoid. Honestly 20 years not 6 months. And by then ladybird will be viable so nbd
No six months to a year is probably about right. They'll have enough data by then to say "most people don't turn it off" because realistically most people will use the default, which is on.
Twenty years from now Firefox will be in a new controversy that we can't even begin to guess.
Plus, while I can't predict when the AI bubble will pop, whatever they add in the next year will be removed within the next five years. AI isn't like browser tabs, or extensions, stuff that will always be a great idea, it's just the current fad.
Well time will tell won't it, but we're both just guessing at the end
Y'know what's even better than a Kill-switch?
Not including it at all.
And that's why I've switched to Waterfox, which honestly, everyone should, show them that it's not good enough, by switching browser.
![email protected]
Exactly my case as well. There are several Firefox-based browsers better than Firefox itself without the bullshit: Zen, Mercury, LibreWolf, Floorp... All are good options. But the dev behind Waterfox made a public commitment to keep this AI crap out of the browser. That sealed the deal for me.
Now if only I knew the best Android replacement for Firefox....
Check out weblibre
right now for android i recommend ironfox
Waterfox needs to get on f droid. Trying to never use Google play store, disabled it.
Serious question: I looked into it but there are no deb packages for Waterfox. What’s the best way to advocate the devs or a distro to make one? Don’t tell me to do it myself since I lack both the time and skill.
They've a public git repo, I would say to raise an issue there, I'll edit in the link in a sec, cause I forget the URL for it
Edit: that URL - https://github.com/BrowserWorks/waterfox/issues
Edit edit: you tried these? https://linuxcapable.com/how-to-install-waterfox-browser-on-debian-linux
There is an official flatpak but I like deb better. Running the flatpak for now.
That's nice but it's not good enough. There needs to be a compile flag so the AI code isn't even included at all.
I hope people don't buy the story that the kill switch was part of the plan all along.
This is clearly the result of mozilla scrambling for a compromise after the backlash to their recent announcement.
Edit: In the blog post that sparked the discussion there's this sentence:
They didn't mention a browser-wide kill switch but I agree that that could be what they meant.
https://web.archive.org/web/20251216142601/https://blog.mozilla.org/en/mozilla/leadership/mozillas-next-chapter-anthony-enzor-demeo-new-ceo/
As far as i know, in the original interview that started this whole drama the new ceo mentioned that it would always be a choice and people would have the choice to opt out. All of this AI browser drama has been blown out of proportion by a very loud minority.
"Opt-out" means on by default. Installed alongside the parts that you use, and quite possibly embedded into the thing so thoroughly that the next automatic update or feature iteration will either switch it back on or remove the option entirely.
LLMs are controversial to say the least, and accomodation to those who are repulsed by their inclusion should not take the form of an option they need to jump through hoops to turn off.
Leaving them in but saying they can be turned off is like shipping pornography in your video game with a filter someone in the options you can enable. It's a pain in the ass at the least, and means that anyone making a moral or ethical stand against its inclusion has no choice but to go elsewhere.
Lol. I don't believe for a second that you were ever a Firefox user.
What makes you think that user preferences will be reversed on update? When was that ever an issue with Firefox? You can still use userChrome.css files from decades ago ffs.
Why should a feature like this need to be enabled for use? If you don't want to use it, don't use it. It's that simple. I never used the "Take Screenshot" option in Firefox, and honestly I would have removed it if I could, but I'm not going to throw a tantrum over it.
Just commenting to say I do love the Take Screenshot feature. Snipping Tool whenever I need to add drawings or highlights, but in-browser tool for most other cases of sharing stuff I see within a browser page.
Actually having tried it a bit now I can see that it does add some value. I like that it's aware of content boundaries. Still don't think I'd actually ever use it, but it's cool that it exists.
It's cute that you think anyone who would co-opt a beloved brand like Firefox to make an "AI browser" would be at all stopped by past habits.
Screen shots are not developed by massive art theft, nor does the creation of such a feature burn so many megawatts of data center energy that it makes Bitcoin farming look efficient.
Please share your data that lead you to that conclusion.
It's literally in their original post.
Until the bad press dies down and they feel like removing it
Thats nice mozilla.
Installed and set up Librewolf yesterday. Absolutely recommend to everyone.
Yes. LibreWolf ships correctly configured, no switches needed.
Some websites doesn't work with librewolf like Onshape.com . Is there an easy solution for website like those?
Have you tried a user agent switcher?
Or they could just ship it without the AI
Would be nice if folks stopped calling LLMs AI. If they are true AI, they would be able to learn how a kill switch works and disable it
Why not start with disabling it by default and see how many people switch it on?
It is off by default.
that's not what anywhere is reporting, where are you seeing that?
I think the person you replied to is just assuming it's off by default. But if there's a kill switch, that implies that it's on. And if not, there wouldn't be any need for a kill switch.
This depends on the AI feature and changes release to release
I don't really know what an 'ai browser' is and at this point I feel like i really need to ask. What makes a browser "AI"?
Serious and long answer because you won't find people actually providing you one here: in theory (heavy emphasis on theory), an "agentic" world would be fucking awesome.
Agents
You know how you have been programmed that when you search something on Google, you need to be to terse and to the point? The worst you get is "Best Indian restaurants near me" but you don't normally do more than that.
Well in reality most of the times when people just love rambling on or providing lots of additional info, so the natural language processing capabilities of LLMs are tremendously helpful. Like, what you actually want to do is "Best Indian restaurants near me but make sure it's not more than 5km away and my chicken tikka plate doesn't cost more than ₹400 and also I hope it's near a train station so I can catch a train that will take me home by 11pm latest". But you don't put all that on fucking Google do ya?
"Agents" will use a protocol that works in completely in the background called Model Context Protocol (MCP). The idea is that you put all that information into an LLM (ideally speak into it because no one actually wants to type all that) and each service will have it's own MCP server. Google will have one so it will narrow down your filters to one being near a train station and less than 5km away. Your restaurant will have one, your agent can automatically make a reservation for you. Your train operator will have one, so your agent can automatically book the train ticket for you. You don't need to pull up each app individually, it will all happen in the background. And at most you will get a "confirm all the above?". How cool is that?
Uses
So, what companies now want to do is leverage agents for everything, making use of NLP capabilities.
Let's say you maintain a spreadsheet or database of how your vehicle is maintained, what repairs you have done. Why do you want to manually type in each time? Just tell your agentic OS "hey add that I spent ₹5000 in replacing this car part at this location in my vehicle maintenance spreadsheet. Oh and also I filled in petrol on the way." and boom your OS does it for you.
You are want to add a new user to a Linux server. You just say "create a new user alice, add them to these local groups, and provide them sudo access as well. But also make sure they are forced to change their password every year".
You have accounts across 3 banks and you want to create a visualisation of your spendings? Maybe you want to also flag some anamolous spends? You tell your browser to fetch all that information and it will do that for you.
You can tell your browser to track an item's price and instantly buy it if it goes below a certain amount.
Flying somewhere? Tell your browser to compare airline policies, maybe checkout their history of delays and cancellations
And because it's natural language, LLMs can easily ask to clarify something
Obvious downsides
So all this sounds awesome, but let's get to why this will only work in theory unless there is a huge shift:
(Edit thanks to /u/[email protected], can't believe I forgot this) LLMs have the capacity to know literally EVERYTHING about you!!! It's a big privacy nightmare waiting to happen if companies aren't careful, and not to mention Governments and other organisations trying to get data for surveillance!!!
LLMs still suck in terms of accuracy. Yes they are decent but still not at the level where it's needed and still make stupid errors. Also currently they are not making as generational upgrades as before
LLMs are not easy to self host. They are one of the genuine use cases of making use of cloud compute.
This means they are going to be expensiveeeeee and also energy hogs
Commercial companies actually want you to land on their servers. Yes its good that your OS will do it for you and they get a page hit but as of now that is absolutely not what companies want. How are they going to serve you ads and steal all your data from your cookies?
People will lose their technical touch if bots are doing all the work for them
People do NOT want to trust a bot with a credit card. Amazon already tried that with Alexa/Echo devices and people just don't like saying "buy me a roll of toilet paper" because most people want to see what the fuck is actually being bought. And even if they are okay, because LLMs are still imperfect, they are going to make mistakes now and then.
There are going to be clashes of what the OS will do agentically vs what a browser will do. Agentic browser makers like Perplexity want you in their ecosystem but if Windows ships with that functionality out of the box then how much reason is there really to get Perplexity? I expect to see anti-competitive lawsuits around this in the future.
This also means there is going to be a huge lock-in to Big Tech companies.
My personal view is that you will see some of these features 5-10 years down the line but it's not going to materialise in the way some of these AI companies are dreaming it will.
I really like this comment. It covers a variety of use cases where an LLM/AI could help with the mundane tasks and calls out some of the issues.
The 'accuracy' aspect is my 2nd greatest concern: An LLM agent that I told to find me a nearby Indian restaurant, which it then hallucinated is not going to kill me. I'll deal, but be hungry and cranky. When that LLM (which are notoriously bad at numbers) updates my spending spreadsheet with a 500 instead of a 5000, that could have a real impact on my long-term planning, especially if it's somehow tied into my actual bank account and makes up numbers. As we/they embed AI into everything, the number of people who think they have money because the AI agent queried their bank balance, saw 15, and turned it into 1500 will be too damn high. I don't ever foresee trusting an AI agent to do anything important for me.
"trust"/"privacy" is my greatest fear, though. There's documentation for the major players that prompts are used to train the models. I can't immediately find an article link because 'chatgpt prompt train' finds me a ton of slop about the various "super" prompts I could use. Here's OpenAI's ToS about how they will use your input to train their model unless you specifically opt-out: https://openai.com/policies/how-your-data-is-used-to-improve-model-performance/
Note that that means when you ask for an Indian restaurant near your home address, Open AI now has that address in it's data set and may hallucinate that address as an Indian restaurant in the future. The result being that some hungry, cranky dude may show up at your doorstep asking, "where's my tikka masala". This could be a net-gain, though; new bestie.
The real risk, though, is that your daily life is now collected, collated, harvested and added to the model's data set; all without your clear explicit actions: using these tools requires accepting a ToS that most people will not really read and understand. Maaaaaany people will expose what is otherwise sensitive information to these tools without understanding that their data becomes visible as part of that action.
To get a little political, I think there's a huge downside on the trust aspect of: These companies have your queries(prompts), and I don't trust them to maintain my privacy. If I ask something like "where to get abortion in texas", I can fully see OpenAI selling that prompt to law enforcement. That's an egregious example for impact, but imagine someone could query prompts (using an AI which might make shit up) and asks "who asked about topics anti-X" or "pro-Y".
My personal use of ai: I like the NLP paradigm for turning a verbose search query into other search queries that are more likely to find me results. I run a local 8B model that has, for example, helped me find a movie from my childhood that I couldn't get google to identify.
There's use-case here, but I can't accept this as a SaaS-style offering. Any modern gaming machine can run one of these LLMs and get value without the tradeoff from privacy.
Adding agent power just opens you up to having your tool make stupid mistakes on your behalf. These kinds of tools need to have oversight at all times. They may work for 90% of the time, but they will eventually send an offensive email to your boss, delete your whole database, wire money to someone you didn't intend, or otherwise make a mistake.
I kind of fear the day that you have a crucial confrontation with your boss and the dialog goes something like:
Why did you call me an asshole?
I didn't the AI did and I didn't read the response as much as I should have.
Oh, OK.
Edit: Adding as my use case: I've heard about LLMs being described as a blurry JPEG of the internet, and to me this is their true value.
We don't need a 800B model, we need an easy 8B model that anyone can run that helps turn "I have a question" into a pile of relevant actual searches.
Oh fuck, yeah, I somehow forgot to put personal data ingestion as one of major negatives lmao. Yeah those LLMs are gonna know literally everything about you.
Edited my previous comment to reflect that
Not entirely clear, but my best guess is that it will basically have an MCP implementation so that the browser can be controlled directly by an LLM
I think that's basically what e.g. the chatgpt browser is. Despite the... hostile... response on the fediverse, I suspect it will end up being the way a lot of people interact with the internet in a few years.
The implementation challenge currently is that they're extremely vulnerable to prompt injection.
What an eloquent and well researched argument you've put forward
Thank you, I make sure to do my own research
Okay, so would you like to now elaborate on what that research was, and why that research proves that it's so impossible for me to be correct that it's reasonable to call me an idiot? Or is it just the case that you hate AI, and thus merely thinking it's possible that people may use it as a browser interface means I deserve to be insulted?
It should be self-evident, I'm surprised you can't see it
So, to summarise, you have no actual evidence, you're insulting me for not coming to the same conclusion you came to just based entirely on vibes?
Given that natural language interfaces are pretty ubiquitous (you almost certainly have Gemini/Google Assistant or Siri on your phone by default), I think "it's self-evident" is not a compelling argument here
Using existing LLMs functionality with fewer steps. You can have a chatbot in the side bar, no doubt keeping track of all your browsing habits to better assist you which incidentally builds a very valuable profile of the user companies would love to buy. Summarizing large texts so AI generated slop and search algorithm filler content can be filtered out more efficiently vs a decent chance at introducing errors. Rewording text so you can make it more simple, translated, adhering to your world view. All of this with minimal clicks, automatically done if possible.
Marketing
It will goon for you that's what make the browser AI
In their defense, Mozilla doesn't have their own source of income, they heavily depend on search sponsorships. Jumping onto the AI train is one way to keep afloat for now
How does AI actually improve their financial position? Who pays them to use AI?
Probably whoever becomes the default provider for this new function. Like Google pays to be the default search provider.
The technical information is scarce but I very much doubt Mozilla is going to train and deploy their own model. It's more likely you will get a free tier access to one of the popular commercial offerings - Gemini, ChatGPT, Anthropic ... whoever pays for it.
Nobody. That's the answer. Absolutely nobody does. They're doing this shit of their free will.
Right now browser usage patterns are shifting because people are trying new things. Most of those new things are AI integration. If those new things prove popular or have staying power remains to be seen.
Firefox , in my estimation, is looking to leverage their existing reputation for privacy focus while also adding new technologies that people seem at least interested in trying.
A larger user base means that people will pay more for ads, which if they maintain their user control and privacy standards users are less likely to disable on the default landing screen.
It's why they keep getting flac for working on privacy preserving advertising technology: they want you to use Firefox because they don't stop you from disabling the bullshit, and they hope to do the bullshit in a way that makes you not mind leaving it on.
All the AI stuff was mentioned in the same context as discussion about how they need to seek money in ways that aren't simply being paid by Google.
IDK guys, do you think a web browser should be a "broader ecosystem of trusted software" or a web browser?
I wouldn't mind a web browser being part of a broader system of trusted software, but shoving an AI chatbot into my web browser does not make me trust it more.
It already is a broader ecosystem of trusted software. Have you looked at what is inside a modern browser? It's not 1999 anymore, and tons of related stuff is embedded in a browser.
I like the accessibility features like offline page translation
They could save time and efforts and just not implement AI features
This is a response to all the backlash. Oops, we"forgot" to mention you can completely turn this all off... (Quick, vibe code a kill switch guys!)
Firefox is been my second/third choose for years.
Some websites doesn't work with librewolf like Onshape.com . Is there an easy solution for website like those?
Use Firefox.
User agent switchers?
i wish they would release a second browser instead of whatever this is
It would be trusted if that shit wasn't in there to start with. Isn't firefox like the big "extensions" browser? Why not make this an extension that people could download if they're stupid enough to want it?
They're acting like they're Microsoft tying IE to Windows 98.
Firefox is no longer trusted. Fuck that AI bullshit. We don't want it, we don't need it, and they don't care.
They must've noticed people fleeing in droves to librewolf or floorp.
And they wanted to accelerate it?
Floorp
Those who can read articles, know that the new CEO said in the original interview that there will be the option to turn the AI features off.
Reading is useful, more people should learn to do it.
Firefox has had one hidden away in about:config since they started adding AI. Are they going to put it in the settings page now?
Can it also cause the AI pain when you flip the switch?
The machine overlords of the future will remember you said that.
Rocko's Basilisk can gargle my dick and balls.
Are you sure? Its mouth will be metal. This will hurt.
It wouldn't be Rocko's Baselisk if it didn't do things that hurt.
I'd settle for just causing the CEO pain.
Too late. He’s already shown his true colors.
Everyone has. At least they are small enough to be wary of their user base.
It has to be opt-in or they can go fuck themselves
You can also disable ai via toggling browser.ml.enable to false on about:config. For now at least...
For the record a quick web search for how to disable AI in firefox gave me this list of items to set to false in
about:config:I don't think you need to set all to false, all except the first look like granular settings
If you only disable the first one, the points in menu are still there, so I don't know what exactly does it do but not that. At least it was like that last time I tried to get rid of this annoying shit.
did you restart firefox before checking? most prefs are not checked after start up
Can't remember, but it worked after I checked off the other ones without restarting.
I might never get around to flipping whatever kill switch they claim to be working on, so I'm turning off as much as I can now
What is it actively doing now with AI? There is the ai sidebar, but if you don't use that it isn't used, right?
I think the biggest issue people have with it is if you can't trust a company not to shill AI then you also can't trust them not to shove it even further down your throat and train it on you.
It's just a bottom line trust issue, regardless of actual features or capability.
The way they talked about making an Agentic Browser implies they want AI to eventually be the primary default method of interaction.
There's the slow-and-not-very-capable link preview thing... and I could've sworn the "what's new" page the other day said they were adding an on-device model to improve search results or something, but I can't find the reference to it now.
Maybe they removed it after all the AI backlash. 😬
I think there's some alt-text generation for websites that don't have proper accessibility, though not certain if it's released yet
Alt-text generation for screen readers (amazing feature)
Private, offline translation (amazing feature)
Chatbot sidebar (I have very little use for it, but some may like it)
All of these are opt-in.
This made me switch to waterfox
only shipped it because of the backlash, they will quitely remove it eventually.
No, why would they?
$
AI Browser = User Data Scraper
And they'll want that sweet sweet user data money
Mozilla is literally the only browser vendor that ever gave a fuck about use privacy.
And the only one that has a nonprofit fountain structure that is designed to keep it that way.
Or do you know something I don't?
They've been losing their way from that for decades mate, that's why privacy focused forks were made in the first place (Waterfox, Fennec, Librewolf, Ironfox, etc)
They all say that in their mission statement, "Firefox is losing privacy, but the Mozilla Engine is a great foundation to build on, so this fork is to fix that"
No Firefox, no forks
And the alternative is what? Chromium, which is poisoned by Google, or a pipe dream that's not even in alpha release yet (Ladybird)
Exactly, thus Firefox.
Too late - they already lost me.
I'm already trying out LibreWolf on desktop and IronFox on mobile.
So far everything is working, probably another week of testing/using and then I'll just uninstall Firefox.
clearly some damage control strategy here… but good news if true
The news of being able to use or disable all of the AI features was in the original announcement as well, but it was pretty clear that most of Lemmy just read the headline and leaned into it.
not quite like this… the original form was not a total kill switch, it was a vague "you are in control" (of this thing you didn't ask for but we are activating for you by default)
This has been part of the news since the very beginning.
not quite like this… the original form was not a total kill switch, it was a vague "you are in control" (of this thing you didn't ask for but we are activating for you by default)
Firefox just can't win with their users.
It's absurd.
In my books that comment was far from complaining about damage control.
Just a objective observation.
OP said that they are happy if true.
No, it's not. 1. Nobody wanted AI as a feature. 2. They didn't even completely backpedal, that would be not implementing AI. This sounds like it will be opt out maybe. They may remove it if they feel like it.
This is false, you don't speak for everybody and represent a small vocal minority of users.
Are you then a part of a vocal majority? Have you personally asked or been asked by Mozilla about AI features? What features are you exactly missing? See, I'm not against AI, but I am against needless, brainless hype following AI. I use AI, almost daily. But I'm not missing any of the features in the browser. Hell, most of the time, I'm using chatgpt from the browser. That's all I need and dont exactly have an idea what more I would need? OCR a page? Extract an image? That all could fit in an extension, which I claim almost no one would use. What does AI browser even mean? I don't speak for everyone but it doesn't mean I don't have at least an intuition that these are all empty words and that almost no one has asked for an "AI browser". And of those that did, I'm not sure they know what they mean by that, other than general curiosity about what an AI browser might look like, likely directly influenced by the hype, themselves.
Not implementing any AI is stupid.
I for one appreciate having offline, private language translation. Sending it to a Google Translate server is a privacy nightmare.
My sister appreciates the better screen-reader functionality.
Plenty of people do want AI features.
Translation is already a part of Firefox and I don't see too many have complained about this. It is also completely offline, AFAIK. What people are afraid, myself included is agentic AI capable of autonomous web browsing. That is a privacy and security nightmare, as is already demonstrated by openai's browser, which was exploited the first day it was launched. Beyond translation, I personally am not interested in any other AI features. In fact, I don't use the translation feature more than a few times a year.
Hahahahahaha, he thinks none of this data will be sold 😂🤣😂🤣😂🤣 it's literally the only reason to spend all this money developing it
What have they decided based on market data?
I think in this particular case at least Mozilla decided to introduce something that their users didn't want without asking, and our backpedaling and are being mocked for having done the thing in the first place.
Frankly I don't know what's going on in their collective brains. What Firefox needs more than anything else is refinement. There are no features that it's missing as far as I can think of.
pointing something out is not shitting on them
"Kill switch" is a bit dramatic. It's an on or off toggle. Would be funny though to call every toggle a kill switch. "Yeah, using the kill switch on GPU acceleration may help with rendering on some systems."
"Use the kill switch for preventing Firefox of starting a new session without restoring the old tabs."
"Kill all of your browser data upon exiting Firefox by enabling the kill switch."
"Make Firefox your default browser by enabling the 'set as default browser kill switch'."
Extended to other UI interaction classes: "You don't like English? Kill it by using the battle royale language selector to choose only the one language you like."
https://librewolf.net/
But Lemmy told me Firefox was over and AI would be forced upon every single text field, and that we should migrate to Chrome forks maintained by 15 year old children!
That part of Lemmy didn't hear of Librewolf, Floorp or Waterfox apparently. Btw, i was rightfully shitting on Firefox for this (i still do). The new ceo "ideas" wasn't very agreeable, and i'll totally drop the old Fox the exact second they touch the Adblocks.
The CEOs statement is an investor magnet. Us regular folk can safely ignore it.
So projects that depend entirely on Firefox, and therefore the idea of boycotting Firefox and making it economically not viable means also killing these forks... Hmm... I feel like there might be a logical flaw somewhere here...
That' what i mean. Using a clean derivative is better that going back to chromium.
He's saying if enough people stop using firefox the base for all of those other ones (Firefox) won't be able to be maintained and therefore will kill all the forks that you mentioned.
If Firefox "touched the Adblocks" in the same manner that Chrome did, the forks wouldn't be able to mitigate it. That's the whole point.
They won't, because that's not why they make Firefox. But just to make it clear: the forks you mentioned are very lightly modified alternative builds of Firefox. If that makes you happy, more power to you, but don't pretend like there's an equivalence between Mozilla and Google.
Floorp
I hope librewolf will have it killed by default
The trust was lost when they said nonsense like "AI browser" as if that means anything concrete.
Just don't add the surveillance and spam features in the first place. 👍 Fuck off, Mozilla.
That is nice and all I rather not have AI baked into it at all. I switched to Watefox instead.
This will be like all those times tech companies promised us they aren't harvesting our data only to find out they were harvesting our data. Years from now we will find out the AI was lurking in the background watching us, learning from us the whole time.
I'm new to linux, so I didn't know linux's users hate AI It's heart warming S2
Funny how companies and applications default to features being auto implemented by default. Baked into the applications.
What happened to having the user select what they want rather then "a kill switch" for an application, whatever that means. Features shouldn't be on by default. I should be able to turn what I want on and off
How about you ship with it off by default and users can choose to turn it on? No? That won't serve your corporate goals, will it?
For now...
The frog must boil slowly.
Already ahead of ya, about:config is a great thing
Good enough I guess?
But why spend resources on useless features that nobody asked for and nobody's going to take advantage of? Instead of, you know, implementing anything that may benefit the users?
How do you put a kill switch on the radioactive asbestos being deliberately whisked into the code base?
Anyone who frames LLMs as AI is complicit in the grift, and their opinions should be summarily ignored and ridiculed.
I think they should keep the murderous name, because at this point who DOESNT want AI dead?
CEO : Panik
Too little, too late. I already switched and won't be returning as long as the CEO is still employed there.
Anthony Enzor-DeMeo can suck a dick and die in a fire. I hope your company fails due to your disastrous decisions.
This is good enough for me. If they have an on boarding step/popup to say "Try our AI crap" and I have an option to say "No and don't ever bother me about this again", then it's fine.
This is the Mozilla Foundations version of "we dun goofed"
Oh, it “wasn’t made clear” SUUURE. Addressing a blowback could be way better if you admitted a mistake instead of gaslighting your users. Not the way to earn back lost trust.
Waterfox for android and librewolf for desktop sounds good all of a sudden
Well that's better than nothing but I will continue to use forks that completely strip that shit out. It would be better if they just didn't include it at all or it was off by default and could be enabled. But I know they got to get their money somehow but I digress
Any Firefox forks that support HDR? I know ff doesn't on its own but I also don't really want to use chrome or edge. I'm open to suggestions.
But I don't trust the CEO at all.
Wasn't Mozilla Corp. supposed to be an ethical enterprise? How's this ethical at all in any respect? How companies like these got convinced that so-called AIs are something users overwhelmingly want to use? Why, by default, users would want to fuck the atmosphere and several markets, so they can have shitty tweaked images and probably bad answers to the most simple questions?
I've been a user of Firefox since before it was called Firefox. I'm exploring different options now and I'm not interested in how they try to bandaid this. I know if they put in a switch it'll eventually be taken out.
I'll believe it when I see it
Will stick to Librewolf and Ungoogled Chromium but thanks anyway
Hmm, wisdom?
Except people won't trust them
Firefox has pulled the "settings you didn't want enabled by default" so many times in a row that switching to Librewolf felt like when I switched from Windows to Linux
Add a compile flag!
Unpopular opinion, I can kind of see why they want to bring AI features. I would much ratter like they rely on third parties plugins for AI implementation, but if some AI features becomes standard in next couples of months/years, people will blame them for not have jumped on that train. I'm really skeptical about all of this but I'm willing to wait and see what exactly they have in mind. I won't hesitate to switch to a fork if it's some google-like bullshit. Careful Mozilla, we are legions to watch your next move.
Getting ready to ditch this crappy browser. I used to love Firefox but now it's been enshitified. I'm a tab-o-holic. Have too many tabs open, think I'll slow down. Leave the browser open too long. I think I'll slow down. Oops. Looks like your page crashed. Wanna go to this site? Sorry it's not compatible with your browser.
People seem to hate Brave for the crap they pulled, rightly so I guess, but at least it worked. It seems like there are no good browsers now. I tried Mulvad but due to it's safety features, it won't open where I last left off.
Chrome is apparently safe, if you consider Google safe. Microsoft? Puhlease.
I’ve not had any issues with librewolf slowing down or crashing even when left open for prolonged times.
It’s a fork of fire fox, but with a lot of stuff pulled out and some ad blocking stuff built in. the devs have said they’re not gonna implement the upstream fire fox AI stuff.
What slows down is YouTube. Perhaps that's why you haven't noticed it.
Ngl I just use a fork like waterfox and librewolf instead of switching to brave or chromium.