This is great news. I recently updated AMD Adrenalin, and the "minimal" version doesn't let you change the distribution of unified RAM on Strix Halo. I installed the "full" version, and it wanted me to install a 10GB "local AI assistant" to "help" me configure it. When I opened the program, it showed me a non-dismissable fake chat that occupied 25% of the screen, prompting me to click it and replace it with a real one.
I remember when every other software prompted you to install Bonzi Buddy or some other intrusive search bar. This AI push is even worse.
Apart from the UI which is crap since their last major update. There are menu options everywhere, two ribbons on the top, a hamburger menu on the right and another on the left. For a long time you opened Thunderbird and it didn't default on the last message that you received but somewhere in the middle of the heap.
> Apart from the UI which is crap since their last major update
But when they updated the UI, they
- Added options to use to make it very close to the old layout
- Set those options for you if you had it customized like that in the previous version
Which is IMHO much better than how Mozilla handled the redesign - you can get the old style in a GitHub repo thanklessly maintained by one person[0], enable userchrome support in about:config (until they decide to take it away one day!), and enable compact mode (also gated behind about:config and called "Compact (not supported)". Oh, and remember to update the userchrome every few updates because they keep breaking it.
That's the difference between user-centric and not user-centric.
It's hard to be in charge of a project like this. You're criticized no matter what you do.
The old UI was criticized by some for being outdated, a mix of old and new styles, didn't fit well with new OS/app styles, etc. It was crap. So they update the UI and it's still crap... for other users. Damned if you do and damned if you don't.
As a longterm thunderbird user I find this annoying. I appreciate it being maintained more actively again but I really liked the fact that the UI stayed stable for years. Changing things to make them "more modern" is just annoying. No one asked for this.
When you delete a message after that the old message remains selected and so if you hit delete again thinking the last time didn’t do the trick, you deleted another message and now go to deleted messages and try to find what you deleted.
The app has a phantom message even in empty folders that it keeps selected. Unread bubble and nothing else, an empty message. You can’t even delete it. Sometimes it persists between app restarts.
It shows unread count on a folder just because it feels like it.
You must be joking. The Supernova UI redesign is an unmitigated disaster. They unnecessarily butchered the look and feel of Thunderbird to the point where people are switching to forks.
Well I like Supernova better, it made me switch to Thunderbird from em Client. (I always wanted to switch to Thunderbird, but until Supernova I never liked the UX enough to keep using it)
The previous UI made no sense. Having the email viewer on the bottom and the list on the top instead of side by side makes absolutely no sense when most emails are designed for viewing in portrait.
What are you talking about? Thunderbird has barely any progress in the last years. It's more busy with breaking and fixing things. Sure, there are reasons for it, but as a user, all I see is stalemate, while one addon after another is dying. Thunderbird Mobile is nice, and I hope Thunderbird Pro will be something good, but so far none of them are the big breakthroughs.
Where are the AI features in Firefox? Looking around right now the only one I see is right click tab -> Summarize page (NEW). I googled a bit and see they have some grouping of tabs feature I've never used/seen (or want). The only other maybe AI feature I remember seeing is the odd left hand bar that is there on fresh installs and I usually remove to declutter.
Are those the features this kill switch removes or was there a deeper issue here?
"- Translations, which help you browse the web in your preferred language.
- Alt text in PDFs, which add accessibility descriptions to images in PDF pages.
- AI-enhanced tab grouping, which suggests related tabs and group names.
- Link previews, which show key points before you open a link.
- AI chatbot in the sidebar, which lets you use your chosen chatbot as you browse, including options like Anthropic Claude, ChatGPT, Microsoft Copilot, Google Gemini and Le Chat Mistral."
I wonder what sort of user testing made them decide that what Firefox users really need is a chat bot in the site bar. Isn't a chat bot in a tab good enough?
And calling translation "AI" seems like deceitful retroactive rebranding. Why is machine translation suddenly "AI" now? It was never branded as such before. Is "AI" here just used to mean machine learning?
The word artificial intelligence was coined in a 1955 research proposal [1] which listed seven aspects of the "artificial intelligence problem". Computers using languages was one of them. Another is "neuron nets", which would indeed encompass a large part of ML and at least Google Translate since circa 2016 [2].
This is also perfectly in line with how the word AI was used until circa 2022. The weird thing is this narrowing of AI to only mean transformer or diffusion based neural network approaches.
And many translation approaches would even fall under that, so not sure how narrow you perceive the term to be now. How do you even define AI to include everything OpenAI calls AI but not include modern translation approaches
IMHO no. Every chatbot has so much wasted space, it really doesn't need to be full-width. Also, what's easier?
Option 1: Being on a tab, copying the URL of the tab, switching to the chatbot tab, pasting the URL and writing some instructions about what to do with that tab.
Option 2: Clicking on the "summarise page" button (whether from the sidebar or from right-click context menu), and having the browser pre-fill the prompt with the URL + the reader view version of the content on that page.
I saw someone else using this feature. It does more than just be a chat bot. You can direct it to automate tasks like go to a web page, search for stuff, etc. -- I asked it to go to pinterest and download the top ten images for "cyberpunk," and it succeeded. Nifty I suppose.
I've run an AI translation tool under a debugger just to see what it did.
It tokenized your input, fed it into a model, then ran the model. Literally the same thing as any other local AI software. Except the model was for translation.
> I wonder what sort of user testing made them decide that what Firefox users really need is a chat bot in the site bar. Isn't a chat bot in a tab good enough?
If it wasn't because I find myself using the AI-sidebar all the time I would probably have shared your opinion.
I guess it's just quite convenient to have it separated from the "regular" tabs and their history.
The transformer architecture that powers large language models was designed by Google for the purposes of machine translation. As others have said, ML and AI have always been closely related if not synonymous.
Wouldn't be the first time. Google gave an option to turn off gemini in Gmail, and suddenly the inbox tabs they had for over a decade decided to disappear.
Firefox's translation specifically is from Project Bergamot: https://browser.mt/ It's a set of language models in the style that people currently call AI.
> AI chatbot in the sidebar, which lets you use your chosen chatbot as you browse, including options like Anthropic Claude, ChatGPT, Microsoft Copilot, Google Gemini and Le Chat Mistral."
When i saw this i expected something more... integrated, but when i tried it with a local LLM (using koboldcpp) after enabling the option to show localhost as an option (it is hidden by default for some reason) all it did was to local whatever webpage was running on the localhost URL (even though koboldcpp also provides an OpenAI compatible endpoint, which is what i expected Firefox to use to provide its own UI). It seems to have some sort of heuristic to find the input box where you type in queries and autofills that with the page text or parts of it (if you have it selected) and that's all.
I kinda expected it instead to use the API endpoint, have its own chat UI, provide MCP tools for accessing and manipulating the page's content, let you create reusable prompts, etc. The current solution feels like something you'd throw together in a weekend at most.
On paper yes. The problem is that they clutter the UI, they trigger at weird times and they turn out to be less useful that they may appear.
Then there's also people, like me, who just want the browser to browse the web. I don't want link preview (annoying feature), Firefox isn't my PDF viewer, I don't have that many tabs that I need to group them and I don't use AI chatbots.
So having a single button to disable all of these features is pretty great. I still want a Firefox Lite, that just does browsing and allows me to add the few extension I want to whatever feature I believe is missing.
I started clicking a 'next page' link before I'd actually finished reading something (so I kept holding the mouse buttown down), and a couple of seconds later Firefox popped up a 'link preview' box informing me that I was clicking on a link to a web forum. Wow, thanks, couldn't have figured that out myself. (It did not actually summarize the next page in any way.)
Page translation is mentioned in TFA. It appears in the address bar on pages detected to be in a foreign language, and is also in the main hamburger menu.
Lol. At first I was thinking it was a AI kill switch on web pages (like Google overview...). I guess was being naive that they would do that, and also weird because there's barely any AI stuff on Firefox indeed...
I switched to LibreWolf, use uBlock Origin and have modified the web search so it uses noai.duckduckgo.com and appends -ai to every Google search query. I'm rather happy with this setup.
That's not a real thing, just a side effect: AI "quick answers" don't appear when you use search operators. In this case, you are using the minus operator to filter results with the word "AI", which might not be desirable.
Good. I was fearing Firefox would also end up having too many AI-Features i do not want. But switching to Chromium-Browsers isnt an option anyways because of their Manifest V3 extension model. Restricting blockers? Whats next?
Only truly independent browser engine left. Firefox is entirely independent on google, but unlike its competitors this dependency is through direct cash payments.
Perhaps the Europeans will end up funding Firefox to get away from American corporate tech dependency. Firefox has tried valiantly to get away from Google money but it's difficult to fund browser development from small donations or without cross-subsidies from other business, like web advertising.
And Brave's built-in blocker uses uBO's and other lists, including allowing trusted filters (which can inject high risk scripts into your sites) automatically. So yes, you are using 3rd party scripts already, and no, they can't monitor all of the commits from all those lists.
Both browsers offer import/export functionality for all of those things. Spend one week trying Firefox and if you don't like it, switch back. The inconvenience here is so small it might as well not exist.
They push millions of lines of code every year, thousands of patches and all kinds of measurable performance improvements so not sure what you're talking about.
Much of the content online is already AI generated, and many websites have their code AI generated, so you'd soon be left with nothing but blank pages. :)
In addition to completely disabling AI, I found the following setting extremely convinent to disable in about:config. They clutter up my right-click on a link or on text selection.
Try modifying a webpage’s source to open with <html mozdisallowselectionprint>. If that works, you can make that universal using any page-mod plugin (though not necessarily the userstyle CSS-only ones). I’m mobile-only tonight so I can’t test myself, but derived from:
`dom.text_fragments.enabled` to remove "Copy link to highlight" on right-click. Yes, it can be useful, but I never use it and the very rare occasion of needing to use it when opening a page, I can just search it myself.
`privacy.query_stripping.strip_on_share.enabled` to remove "Copy clean link". I would rather it just did that clean link thing automatically, but I don't actually care about clean links -- it's just annoying having two "copy link" next to each other (especially with one which is greyed out 99% of the time!)
It strips query params used for tracking. Has it never bothered you when you click share on Instagram or YouTube and it inserts a unique-to-you "share ID" in the URL? I was burned by this once, now every time I copy a share URL from a social media platform I first paste it into my text editor and remove all of the privacy invasive tracking cruft.
As an aside I think it's only matter of time before this is done without query params and instead each share link is generated just for you.
I wish we could also disable "Send via email" when right clicking a picture, I constantly misclick and userChrome.css does not work for the context menu on macOS since it's natively rendered...
Ironically, I bet that a significant majority of the users that turn on the AI kill switch — which must have some kind of phone-home telematics attached — will also be users who have disabled Firefox metrics collection and so will not have their opinion counted.
So, the most effective path here for y’all to be heard is not flipping the switch off yourself (do so anyways!) — anyone who cares at this stage has probably opted out of being counted already, after all — but instead to ensure that news of this switch spreads to absolutely as many non-tech people as possible. Don’t argue that they should run some script that shuts off their metrics and phone home and updates. Just convince them to shut off the AI and explain that this is why their browser got slow about a year ago! They’ll flip off the switch gleefully, their phone-home will count them, and y’all will have the strongest possible impact on the telematics graphs at Mozilla.
I already ran the disable process manually on the computers I have friends and family IT duties towards, so I’ll go back and do the AI switch to be sure it’s counted next week. Yes, this is a crap way to be heard. But making a mark on feature opt-out graphs is probably the only hope we have left to get their executive leadership to stop drowning the browser for its own good.
The other thing people can do is install Firefox and use it. An uptick in user share also serves as a metric to reinforce the move. Let's be honest, most people complaining are using chrome or some flavor.
But current Firefox users could probably temporarily turn on telemetry, activate the kill switch, and turn telemetry back off. Just make sure you wait long enough to ensure the information is sent
They should get the best metrics out of update server communications anyhow, 30% or more users getting update downloads than having AI enabled should be obvious.
I've read this three times now and still don't understand you. Do you think people don't update their software? Are you just counting every update as people happy about the kill switch?
There'll be so much noise in that signal it'll be almost useless. You can't differentiate it from anything else. For all anyone knows it happened because the word Firefox was in the top story on hacker news
If my choice is between a single blatant signal of hostility to AI that can’t be misunderstood, and hoping that a pro-AI company’s executives invested in voluntarily correlating two different sets of logs to prove itself wrong, then I’m taking door number one.
> will also be users who have disabled Firefox metrics collection and so will not have their opinion counted.
Gee. If only there was a way to collect users opinions on things. Welp.. guess we have to live with subtly spying on everything they do with our software.
Most people who are vocal aren't representative of users.
Many vocal people aren't even users.
Don't get me wrong, I turn off telemetry, but you're acting like it's easy to get that information. You act like people don't scream when Firefox prompts people with surveys. You act like there isn't bias in survey takers.
If you just pretend everything is easy we'll just end up reinventing the same evils we're trying to fight today. Unfortunately most evils are created from good intentions. I hear there's an entire road paved that way
> but you're acting like it's easy to get that information.
It is, relatively speaking.
> You act like people don't scream when Firefox prompts people with surveys.
Surveys without proper response and adjustments aren't passing feedback, it's political theatre. People groan about surveys because it takes time and rarely shows results reflectant of the responses.
We know the system is broken. Hard to shame us into thinking we're the ones who broke it.
> You act like there isn't bias in survey takers.
You act like statisticians don't spend half their field accounting around bias.
>we'll just end up reinventing the same evils we're trying to fight today.
Let's have the old evils dealt with before worrying about creating new evils who happen to do the exact same thing as the "old evils" (spoilers: they are the same picture).
People who aren't vocal tend not to care very much. People aren't vocal tend to just accept whatever happens to them. So why not cater to the people who are vocal?
If we replaced telemetry with some sort of survey emails and phone calls, we'd get exactly another 500-thread discussion on HN about how "Mozilla is collecting emails to sell to the highest bidder!", "Mozilla is sending us spam!" and whatnot.
I think at this point they know all these opinions pretty well, but they simply don't care or see better growth options by targeting users who don't belong to that particular bubble. They see OpenAI approaching the same active user counts as Facebook and they want a slice of that pie. And the majority of that pie is non-techies.
Arguing that telemetry is wrong doesn’t seem to have stopped Firefox from using it. If your fight against telemetry is a higher priority than your fight against AI in the browser, good luck and more power to you! I’m just making sure no one makes that decision by accident.
This isn’t about a fight between two things. This is about a company that completely refuses to listen to their users and then some random guy on the Internet who says you should deal with their telemetry to somehow fix the problem that in no way relates to the actual problem and no way has evidence that they would take steps to fix the problem either. It’s very prosocial to think that Firefox actually makes changes based on user feedback, but almost none of these features have users been clamoring for, this is totally executive bullshit pushed into the product.
The best thing about Firefox telemetry is that it can't be easily disabled. There are many setting that control it. Including an external scheduled task that can't be disabled using Firefox itself. And even if you delete the task it will come back after update.
I don't see the appeal, it takes more "clicks" to do many actions and I had to disable the ridiculous new oversized "rectangle tab preview block" (whatever it's called).
yes exactly, their design was already better than chrome and condensed but now we have these outdated round and padding heavy toy controls again, just why?
I actually liked the previous UI much more, the new one looks like a baby toy and uses more space because of the control padding. completely unnecessary.
This UI is great but do you get this horrible thing where sometimes the browser is shows a white screen and you have to force stop the app? Happens all the time on the latest version for my Pixel 9a. And did on my Pixel 7 too before. It's really horrible and I can't pin down any rhythm or reason other than loosely seeming to happen more often when I'm in battery saver mode.
I don't have that on fenec and ironfox on my pixel 6a running graphene.
What I have always had with any firefox based browser on android is erratic behavior in text field. Most of the time it works well but sometimes on some commenting systems my input is duplicated/multiplicated/garbled, trying to select where you want to insert words in the middle of the sentence sometimes becomes impossible, always resetting to the beginning of the text field, etc.
On some websites it only rarely happens, on some it is much more regular. Never understod why but when I want to edit a comment I have to resort on a regular basis to copy the full comment to a note app, edit the text, and replace it in its entirety in the text field.
Super annoying but still less annoying than using a chrome based browser with no way to remove ads and have a bit of privacy control.
Firefox does what some people want, people complain. Firefox does what other people want, people complain. Firefox does what both people want, people complain.
I'm sorry, but we'll never get corporations to do what we want if we don't throw them the smallest bone when we get our way. You need positive reinforcement too, not just negative. If it's all negative they just stop caring and you get companies lot Google who just don't give a shit anymore.
And yes, there are some AI features I like and I want in the browser. I get a lot of utility out of translation as well as semantic search of my history. I don't want agents in my browser but get, Firefox is giving us choices.
Look, no one needs to like Firefox, but let's also be honest, it's the best we got right now. Google, Apple, and Microsoft are shoving agents down our throats and putting us in walled gardens that are getting harder and harder to break from. I don't care what flavor of chromium you use, Google is still using it to control the way the web works. Everyone loves to say how chromium is has greater coverage of standards but never takes a second to question who sets those standards.
I'm sorry guys, that's the state of things now. You can't fight Google by switching to chromium. It's still their vehicle to eat the internet. Our choices right now are Safari, Firefox, and maybe ladybird. It's slim pickings and nothing is close to perfect. At this point it doesn't even matter if Mozilla is evil, because at least they're the enemy of our enemy. Google is keeping them on life support to avoid monopoly claims but how long will they need that?
So what, we're just going to hand the keys of the kingdom to the guys selling artisian turd sandwiches because what, there isn't enough mayo on your ham sandwich? Because you don't like ham?
We got a win. Celebrate. Take the break from being cynical. There's bigger battles to fight and there'll be more tomorrow. Take the night off and don't be a sore winner
> So what, we're just going to hand the keys of the kingdom to the guys selling artisian turd sandwiches because what, there isn't enough mayo on your ham sandwich? Because you don't like ham?
Firefox is the artisan turd sandwich. They are burning dev time on features barely anyone asked, while bleeding market share for last decade
This is a narrative that got hallucinated into existence by comment sections. Most of the effort goes towards the core browser and always has, side bets have found their ways into headlines and comment sections from users overestimating the resources necessary to build them out while underestimating the scale of regular work like security patches and performance improvements.
Who has expressed a desire for Firefox to become "an AI browser"?
Because that's the source of the complaints. I don't want to use an "AI browser", kill switch or not. If this "AI browser" dies because of their mission to destroy community goodwill, good. I'm sick of giving the benefit of the doubt every time they royally fuck up. This situation where they're the steward of the only non-Google browser is not tenable, something needs to change.
Is it so hard to understand that the features person A wants are not necessarily the same as person B yet a company/foundation would be interested in providing a complete featureset for both A and B?
I don't need themes nor having my url bar serving search. I am not interested in an AI agent in my browser yet I welcome traduction features, do I have to shit on every company developping a software that has some features because I don't want them?
I am much more pragmatic: are these features easy to ignore/disable, do they largely increase the resources needed (disk, memory, cpu) even when not used, do these introduce bloat, etc.
I wasn't interested in pocket, I was just using a combination of firefox forks or disabling it on the devices I was using. That is the whole point of open source software.
> This situation where they're the steward of the only non-Google browser is not tenable, something needs to change.
So what, you're going to help Google shove the knife in deeper? Idk man, seems like a bad way to fight Google.
But honestly it just feels like you didn't even read my comment. I'm sorry that it's a lot, but I'm petty sure people can handle 10-30 seconds of reading. I even said it doesn't matter if Mozilla is evil. How do you turn that into me giving them the benefit of the doubt. I'm literally just arguing that there's slim pickings and to not help our bigger enemy to kill their enemies. It doesn't matter if their enemies ate evil, you're just helping the bigger evil get bigger and consolidate power. I'm saying "there's more important problems right now, not be fucking dumb and get distracted or before you know it you'll lose your head"
Letting Mozilla torpedo the only non-Google browser is also a bad way to fight Google. It's looking bleak. The only hope is that Mozilla dies and someone more serious picks up the mantle. Not sure who that would be though.
If I smash your window every day but one day after being asked to stop again and again I decide to just leave poop on your front porch instead, should you not complain about the poop?
Malicious compliance is no compliance, it is still malicious.
Firefox already captured the developer audience, and it wasn't enough. So they pivoted to whatever they thought of to increase mainstream adoption. That is really alienating that developer audience though.
Firefox had mainstream adoption in the Firefox 2 era, before Chrome launched. What drew people to Chrome was that it was fast, efficient, and did I mention fast? Unfortunately all browsers today seem to have forgotten how important that is.
The whole purpose of Chrome is to have google search by default in people's browsers. Once they have a large market share nothing else really matters. If people truly cared about speed Google would have changed it already.
Yes, I really like it. I wish they let extensions access it directly but it seems they don't (yet, at least). I'm also a bit surprised more hasn't been done to distribute it separately as a command line tool as I think something like that would be very helpful. I have seen a couple of attempts to put a CLI wrapper around it but they are third-party and seem to not be well maintained or documented.
I wish there were some updates about PWA support.
Haven’t heard about progress on this since last August.
Is it still in beta and only available on Windows?
At this point, the industry wide decision to have AI as opt-out is suspicious. Why not have an AI opt-in default? It would certainly save so much trouble from the Luddites. On the other hand, an AI opt-in default would be not using any technology at all (where opting in would be turning on your computer or device) under a dictionary definition of artificial intelligence.
I'm torn on whether to see this "AI Kill switch" as a win on respecting the users, or something to keep us distractewd while they ship through "Trusted Types" API that sounds like further restriction of end-user computing freedoms.
I would absolutely love to hear your reasoning that leads to type systems being considered a "restriction of end-user computing freedoms".
For those that don't know what trusted types are: Simply put, it splits the string type in to unsanitised_string_from_user and safe_escaped_string where unsafe strings can not be used in function parameters that only take a safe string That's heavily simplifying of course, but it's the basic idea.
Skimming the API docs on MDN, it makes sure the site vendor gets to run filtering code over anything you'd want to inject via e.g. user script or console, securing it with CSP. I expect this to make user scripts work as well as they do on Chrome now. If there's a workaround, I'd love to hear about it.
You may be thinking of the much-hated "Trusted Computing" initiative. "Trusted" here means that the JavaScript dev picks a sanitizing library they trust, not that Mozilla decides what software is trustworthy.
Nah, my issue isn't with users vs. Mozilla, but users vs. "JavaScript dev", specifically the difference of opinion on who should have final say on what gets executed and what doesn't.
Question is, can you sidestep or disable them in user scripts or in developer tools, without disabling CSP entirely or doing something even more invasive (and generally precluding use of that browser instance for browsing)?
Before, we did not need to disable AI stuff. Now Mozilla forced us (that is those of us who don't like or use AI) into an extra step. Guess the only thing worse is being given no choice at all though.
It is such a shame where Firefox is and is headed these days. It could have been a browser for the people, instead it's an advertising machine, similar to Chrome.
last time when I updated Firefox, the package manager began building ONNX Runtime from source, which my "minuscule" 16GB of RAM couldn't handle. I want that during install time, as I don't like the idea of rebuilding ONNX every time Firefox updates, period.
no, NixOS Unstable. normally, they precompile such stuff, but ONNX decided that it wants to link against my ROCm instance specifically. looks like soon we will have to resort to dirty workarounds akin to those in Blender (HIP, CUDA support)
I don’t want nor need on device translation enabled by default. I’ve gone without it for the three decades in which browsers have been around. I’m sure it’s brilliantly useful for some people. A one time ‘would you like to enable AI on startup’ for at least years with user profiles that are significantly old would at least be a show of good faith.
It's more like if eating guano became a fad, and restaurants started offering guano dishes, and this one said "OK we'll still have guano on the menu because it is popular with some patrons, but we'll have a separate menu with no guano options at all for those who want to be sure they're not getting any."
> This is like a restaurant that releases a new feature that they will no longer defecate in your food.
Thank god, at least there's one restaurant not serving literal shit.
You're analogy works but you can't forget that there other restaurants. That the other restaurant not only aren't making promises to not defecate in your for but they're actively advertising how much shit they can shove in a sandwich. Even the bread is made of shit!
So thank fucking god. At least there's one place where I don't have to eat shit. The bar is so fucking low it doesn't matter if they spit in it or you find the chef's ball hairs, at least it isn't shit.
"The 'AI Kill Switch' in Firefox 148 is not just a feature; it's an architectural statement on Agency. Most platforms today operate on a 'Seed-to-Harvest' model where the user is merely the soil for data collection. By allowing a hard-disable of AI functionalities, we are seeing the emergence of a 'Watchman' architecture—one that prioritizes the user's right to define 'Value' beyond algorithmic efficiency. We need more systems that respect the 'Seed' (the origin of choice) rather than just maximizing the output."
Maybe it's just me but why is it on by default? Why is this shit not off in the first place? Why can't Firefox just be a browser with great html, css, js rendering and then have a bunch of toggles for extra crap that people want? Do they actually have metrics that show "When we enable this crap by default we make an extra $X Million per year"? I'd put money on that being untrue. I bet it's like the data-driven ad spend - I've yet to see anything that proves that hoovering up bajillions of data points on each person moves the needle on spending beyond just showing a context-relevant ad, e.g. An ad for fishing gear on a blog post about fishing gear!
Honestly, I feel more and more every day like old-man-shaking-fist-at-clouds! Can we not just have something that works without spying, without engagement-driven shit switched on all the fucking time?
I think of the Simpsons Mr Brown meme where he's asking "Is it me that's wrong?".
Because the vast majority of Firefox users aren't contrarians or diehard anti AI crusaders. These features are helpful to people. So they're on and available for people to use, and since it's a foss customisable browser, users with strong opinions can change it to their liking.
I don't think the AI features are used for advertising data harvesting but I'm happy to be educated if you have a source saying otherwise.
>Maybe it's just me but why is it on by default? Why is this shit not off in the first place?
Because there would be millions of emails to support asking "why can't Firefox translate a page the same way chrome can?" from people that couldn't find the AI opt-in switch.
> Can we not just have something that works without spying, without engagement-driven shit switched on all the fucking time?
Absolutely not! Making someone opt-in (the horror!) would result in too much confusion and support tickets! We can't ignore those tickets and tell people to figure it out, we have to cave and turn all the bullshit on by default.
I think a lot of our problems can be explained by: corporations lacking a spine and people unwilling to learn about the computer they use everyday. It's basically unacceptable to not know how to change a tire if you own a car but clicking through some browser settings is too hard, I guess.
Expecting everyone to learn a lot about Firefox to turn on features they want is a lot more of a burden on the users than expecting opinionated power users to remove features they don't want.
(This post is directed to all software that shoves features like this in my face, and especially Microsoft more than Firefox.)
My problem with all software that shoves these AI features in my face, is that I don’t use features under duress.
If you interrupt what I’m doing to push me to use a feature, I won’t use it. If you’re a web designer and you block the page to tell me to sign up for an account, I close the tab and vow to never create an account. If you stop what I’m doing to ask me to rate your app, I’m going to give it 1 star. Et cetera.
Now I’ll be the first to admit this is childish… it’s a flaw in my character. When I feel pushed, I push back, and software pushing me makes me irrationally angry for reasons I can’t quite articulate. In some ways I wish I wasn’t like this. But I can’t be alone. I’m certain there is a non-negligible number of people like me, and when a browser immediately shoves AI features in my face on first launch, well, the first thing I’m going to do is disable them.
The especially tragic part is that I personally find LLMs useful! And I’m at the point where I sorta want to install a Firefox extension for ChatGPT now. But the actual browser AI features were pushed on me in a way that made me feel violated, so I can’t use them on principle. Maybe in a few years I guess.
If instead these companies would just dial it back several notches, I would have had the curiosity to try these features out myself, and I’d likely be using them by now. But the way they’ve tried so hard to force them on me has destroyed my trust and now, not only am I not using whatever feature they promote, I hate their product more than I otherwise would.
Firefox isn’t actually that bad here, and now that there’s a simple kill switch, I may actually try their chatbot sidebar thing. But for companies like MS, I will never, ever, ever use any of their AI features for the reasons above. (I’ve literally uninstalled Windows now, it’s gotten so bad.)
How is a company supposed to accommodate your personality while also showing other user new features that Firefox just added to the browser? If you're a contrarian by nature Firefox can't help you nor change their strategy to suit you, because the majority of users aren't like you.
You never, ever need to interrupt the user or stop them from what they’re doing in order to show them some new feature.
Here’s what you do:
- Make a “what’s new” section in settings.
- Put a link in there that takes me to a webpage where I can see what’s new.
- That’s it.
Instead what always seems to happen is that I’m in the middle of trying to do something with the software and in order to do that I must close whatever popup you’re shoving in my face to tell me about the new feature. I don’t have time to read it now because I opened the software with an intention to finish a task and I don’t have time to read it now. And then later when I finally do have the time to look at your new feature? Nobody bothers actually putting that information anywhere persistent, so I guess I’m out of luck even if I care about the feature.
Updating to a new iOS version is a perfect example of this. Say there are a dozen apps that have a new feature popup on first launch when you update iOS. Imagine a typical day waking up and trying to use my phone. I have to drive somewhere so I try to put an address into Maps and have to immediately fend off the “what’s new in maps” dialog so I can type the address I need to go to. Then I want to put on the song my kid is yelling at me to play and have to fend off the “what’s new in Music” popup. Later I’m trying to respond to an important text and have to fend off “What’s new in messages”, etc etc etc.
That first day using iOS is an absolute nightmare because of this.
Now imagine the alternative: a simple badge icon in the settings app, and I tap it and see a link to “what’s new in iOS”, and guess what, it can be a fucking webpage! I can bookmark it! I can add it to my reading list and see it later! Hell, I can even share it with my friends!
But no, instead apps insist on trying to increase “engagement” with their new feature, because some PM’s promo packet wants to include “this many users used my new feature” and the only way they can think to do this is to (1) stop the user from accomplishing their task until they tap the cutesy “Got It!” button, and (2) don’t bother with persisting it anywhere, because the idea that the user doesn’t have time to check out the feature now is so foreign to these sociopaths it never even crossed their mind.
Whatever works for large numbers is what will happen.
But overall, you and I (and many) will try to push back and insist on consent.
The sign-up form with an unchecked "sign me up for your newsletter" option.
The first-run experience with a question... "do you want us to notify you of new features?"
But this is not the norm, and even if good actors get rewarded by a few childish customers, bad actors seem to get rewarded much more by a massive infusion of funds.
Why wasn't this there from the get go? Many people dislike the AI spam; I do too. I use chrome-based browsers usually (I also hate how dependent I have become on Google; default firefox refuses to play audio on my linux system as they claim we need pulseaudio, chrome instead makes no such assumption and audio plays just fine, so one can go and figure out why mozilla acts worse than Google here - all the google-bribe money killed its THINKING ability), so when I do, I use a few extensions such as "disable AI overview" or similar. It is annoying that we have to invest time in order to uncripple the world wide web. Browser vendors should be much more responsible, from the get go. But they all want to jump on the hype train, to milk out more money. Greed is the driving theme nowadays. (They could offer AI based on people who want or need that, rather than cram it down onto everyone.)
Quite a few of the LLM features actually add value for a certain group of users. Automated image descriptions for the visually impaired, automatic translation, ... Running those on local models is a net benefit for quite a few people, but they get a bad rep because they're "AI" and the current trend of shoving AI everyplace and with no means of escape means that AI in general has a - well deserved - bad reputation.
It's basically ungoogled-chromium with manifest v2 support. Chromium is just technically superior than Firefox. It's a simple fact. The problem is the telemetry and AI features they added in it, which Helium or ungoogled-chromium doesn't have.
Firefox is the only holdout against the ad companies, and I'm counting Microsoft amongst those. It's a very good browser, independent with its own renderer, with decent ad blocking and decent performance.
It continually amazes me how people use a Google product on their desktop, as if they don't send enough data to an ad company. Actually, I'm not sure why I type this, any rational arguments are definitely not winning them over.
In theory you are not incorrect, but Google bribes Firefox and Google makes most money via ads. Mozilla gave up on firefox a long time ago.
> It continually amazes me how people use a Google product on their desktop, as if they don't send enough data to an ad company.
I'd love to have alternatives, but which ones are there? Firefox is not an alternative; audio does not work for me as I am pulseaudio free here. On chrome-based browsers audio works fine, out of the box, so it is not my system that is at fault; it is mozilla that is at fault. I also reported this, the lazy firefox dev said all Linux users use pulseuaudio these days. Well ...
I could recompile it but compiling firefox is a pain in the ...:
I am not going to use a build system that is +20 years old and only exists because Mozilla is too lazy to switch to cmake or meson/ninja as primary build tool.
> Actually, I'm not sure why I type this, any rational arguments are definitely not winning them over.
Well I gave one rational argument: can't play audio on my linux box if I use firefox (by default that is). I can give many more reasons too. You seem to make the point that Google is worse, so we should also use a bad product (firefox). I think we really need better browsers in general. Firefox simply isn't one and that is Mozilla's fault. There is a reason why it went into decline. Mozilla gave up the fight - the ad-money made it weak.
> Firefox is not an alternative; audio does not work for me. I could recompile it but compiling firefox is a pain in the ...
Obviously I don't have any data backing me up here, but I'm going to guess that that isn't the main reason why so many people choose Chrome over Firefox.
Firefox has been my main browser lately, and in my experience it covers pretty much every latest spec: no issues with Web Audio, WebGL (as well as WebGPU, I think), CSS features, etc. There are some select cases where Chrome has deployed something and Firefox is lagging (Background Fetch, for example) but that affects me more as a developer than a user. I cannot remember a single time when I opened something and it didn’t work in Firefox.
> I could recompile it but compiling firefox is a pain in the ...
Would second this. Mach uses Python, and the dependencies they use are a pain whenever no pre-built wheels are available. Especially so when you see that an "optional" Mach dependency for build system telemetry is what busting the configuration (not build) stage...
Do you mean you disable pipewire-pulse? Why? Or does audio not work for you with pipewire-pulse? I've never had issues with firefox and pipewire-pulse on my system.
I daily drive FF in desktop and Android but Brave has doubled in users the last few years, and my mildly tech-conscious acquaintances have settled on it after Manifest v3, while FF has been flat. That has been the greatest vote of no confidence against it ever.
You're being down-voted because it's a low-effort comment which comes with a large burden of proof that you've not included. Specifically:
> mozilla is basically a google subsidiary
"Everyone" knows that Mozilla has a heavy financial reliance on Google. So are you bringing this up to suggest that Mozilla also consistently acts to benefit Google and its ad network? If so, where's the proof? If not, what's the point you're making?
> firefox telemetry is almost comparable to chrome
Comparable to Chrome what? Telemetry? Something else? What is Firefox using that data for? In the service of or against users? What's the point you're trying to make? If you're making assertions, where's the proof?
You're making a lot of imprecise comments, most of interpretations of which carry a large burden of proof, and then complaining that people are just down-voting and moving on.
Mozilla Corporation takes money from Google for search placement. That doesn’t turn it into a subsidiary. Google doesn’t own it, doesn’t run its roadmap, and doesn’t ship its code. Mozilla negotiates search deals the same way Apple does for Safari. Revenue deal ≠ corporate control.
On telemetry: you’re overstating it. Firefox ships with telemetry on, but it documents what it collects, lets users turn it off, and exposes most of it in about:config. Google Chrome ties into a much broader account system, sync stack, and ad network. Chrome doesn’t operate in isolation; it plugs straight into Google’s data ecosystem. Firefox doesn’t own an ad network to feed.
“Almost comparable” needs evidence. Comparable how? Volume? Type? Identifiability? Retention? Without specifics, the claim collapses into vibes.
The bigger difference sits lower in the stack: engine independence. Firefox runs on Gecko. Chrome runs on Blink. If you care about web monoculture, that matters more than marginal telemetry deltas. When one engine dominates, web standards start drifting toward what that engine implements. We watched that happen in the IE6 era.
As for uBlock Origin: yes, it’s a major reason people choose Firefox. But browser architecture shapes how long powerful content blockers survive. Chrome’s extension model changes (Manifest V3) restrict what blockers can do. Firefox kept the older, more capable API. That choice signals priorities.
If your argument reduces to “both collect some data, so it doesn’t matter,” you flatten meaningful differences. The question isn’t purity. The question asks who controls the engine, who sets extension policy, and who benefits from surveillance at scale.
If you think those differences don’t matter, make that case directly. But don’t blur structural distinctions into “basically the same.” They’re not.
Thing is, there's a large (or at least certainly vocal) contingent
of users (and mostly techies, to boot) that view "AI" as the Devil, and transformer models as the original sin, and they want to refuse to partake, wholesale.
This feature seems to be a nod to people with this worldview.
I figure, hey, at least Mozilla listened and provided the opt-out. It could be worse. I also happen to be in the "food defecating analogy" camp, and I can give the developer an unenthusiastic thumbs up for at least listening to the peanut gallery this time.
Ideally they wouldn't make the product bad, with a badness opt-out, in the first place, but everyone in Silicon Valley's got to feed the AI monkey. So I guess this is the best we can expect.
FWIW, I may be in the other camp but I strongly respect them for providing this feature. It perhaps wouldn't be necessary if the pro-AI push wasn't so ham-fisted and utterly disrespectful of users for the past years.
(Also I didn't realize how bad this push got until I visited California recently, and saw every other billboard - that's physical ad over a road - pushing some unqualified form of AI magic on me).
I use AI for coding, I don't want it on my browser. Actually I don't want it on Firefox. Let Brave be the AI browser. They could have used AI themselves to actually make their pathetic new split tab view usable.
> For those who wish to maintain some AI functionalities, a selective blocking option is available, enabling users to retain useful features like on-device translations while avoiding cloud-based services.
Sure, but then the people complaining would need to recognize that that's not happening. All the AI features are local models. The only thing not local is that side window you can open that can connect to a chatbot. It's also easy to disable.
Honestly, I think people just like complaining. I think they like complaining about Firefox even more. There's plenty to complain about, but aren't there bigger fish to fry right now? Seems like complaining about some minnows while we're being circled by a bunch of great whites
I think a lot of people are simply sick of seeing "HEY NEW AI FEATURE LOOK AT ME" popups everywhere. Shove something into people's faces often enough and it becomes like a red rag to a bull. That's probably not the only thing, but it's one big motivation why some people want it.
People vehemently asked for a kill switch that does exactly that - kill off all AI-related features. I quite like the local LLM translations etc., but jedem Tierchen sein Plaisierchen, as they say over here.
Seems like willful disregard of the fact that "AI" has been now pushed as synonymous with chatbots, "AI summaries", LLM text generation when convenient, but then at criticism, it goes back to "oh but machine translators are AI too".
All:
(1) Generated comments aren't allowed on HN - this rule predates LLMs but obviously applies even more now: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&query=by%3Adang%20%22generated%20comments%22&sort=byDate&type=comment
(2) If you see accounts that look like they're mostly posting genAI comments, please let us know at hn@ycombinator.com. That's how I found my way to these cases.
>"But I use it to help my English!!!! Who cares if it's AI if the comment is good??????????"
Please don't post generated or AI-filtered posts to HN. We want to hear you in your own voice, and it's fine if your English isn't perfect.
If you don't flag this shit when you see it, HN is fucked. The commons is fucked. And you all keep upvoting and replying to comments from accounts that start posting 30 comments a day after 2 years of silence that are all one tightly packed paragraph of pablum with the same structure and tells that are harder to fix than replacing the em dash with a double hyphen.
---
P.S. The actual comment is (surprise!) completely wrong. Visit Settings -> AI Controls and you will see a granular set of feature switches under the master kill switch. Each has a clear title and description and is independent.
I don't think AI features in a browser are bad, and I think people who tut-tut it are overboard.
However, I think data control is critical and any kind of implicit cloud service such as transmission to remote AI servers should be toggle-able clearly, just like search autocomplete can be done.
I was being sincere, my friend. I genuinely envy that worldview. I long for it. I wish above nearly all else I could reset. I was being authentic and vulnerable. Why does that infuriate you?
Read your statement again and tell me who the pessimist is, and who is most in need of a break from the internet.
The legal structures that mandate what power google actually has over mozilla still presumably exist though. Pretending that we are in full blown dictatorship is, in my view, still cynical.
Though of course there’s no telling how far we will eventually go in a trumpworld.
> Pretending that we are in full blown dictatorship is, in my view, still cynical.
Could you please point to what I said that implies I'm pretending a "full blown dictatorship?" I apologize if that's somehow what I indicated. It certainly wasn't my intent.
i intended to ask what the difference was between two browsers that are both beholden to a company whose express goal is to suck up personal data. so far ive gotten vitriol, AI, and downvotes. my actual question remains unanswered. if you'd like to answer the question that would be cool! but yeah if you dont want to answer , it'd probably be easier to say nothing than to tell me to die alone
What an absolute dystopia when disabling AI is considered a "feature". Obviously the _enabling_ of it should be the "feature" if anything, and considering popular opinion there is absolutely no reason to enable it by default.
> considering popular opinion there is absolutely no reason to enable it by default
You need to update your priors. The popular opinion is there is a reason to enable it by default. ChatGPT is the #5 most popular website in the world, more popular than Wikipedia, Reddit or Twitter. The vast majority of users want to use AI.
Here's a Firefox file [1] specifically for integrating YouTube videos into their picture-in-picture system. Your random video website won't get this treatment of course, need to be a popular one.
Here's a piece of C++ code [2] in the Firefox engine that specifically rewrites old YouTube embeds from their old HTML embed snippet to the new one. Again, your own video website will never be so deeply integrated into Firefox because it's not a top 10 website.
Firefox Readability mode makes pages more readable by removing useless stuff like videos. Unless it's a YouTube [3] or other top-N popular video website of course. YouTube videos are given special treatment because it's popular and having small integrations like this make the user experience better.
I don't mind the AI features per se, but is there a configuration setting to sent the traffic through a local AI Gateway to prevent the AI from receiving private information? At the very least to track what is sent over the wire.
I re-tried Firefox couple of weeks ago, in the latest episode of a series of tries to "finally" migrating to it. Same CPU fans blowing like jet engines randomly. None of that with Vivaldi (which is anyway all Chromium/Blink underneath) - so came back to it.
Firefox has so many nice things like containers but basic performance issues are still unresolved.
I think that depends on what sites you are using/things you are doing with it. The only time the fans turn on for me is one specific dashboard in Home Assistant. Nothing else I connect to even moves the needle on temperature/CPU use
The page seems to be a copy from the original Mozilla press release from February 2nd: https://blog.mozilla.org/en/firefox/ai-controls/
It was discussed here: https://news.ycombinator.com/item?id=46858492
Yeah flagged as a spammy source.
This is great news. I recently updated AMD Adrenalin, and the "minimal" version doesn't let you change the distribution of unified RAM on Strix Halo. I installed the "full" version, and it wanted me to install a 10GB "local AI assistant" to "help" me configure it. When I opened the program, it showed me a non-dismissable fake chat that occupied 25% of the screen, prompting me to click it and replace it with a real one.
I remember when every other software prompted you to install Bonzi Buddy or some other intrusive search bar. This AI push is even worse.
Please let whoever steers Thunderbird development and road map also steer Firefox.
Thunderbird is at the moment the pinnacle of user-centered, focused and down-to-earth development of open-source software.
That’s because Thunderbird is no longer part of Mozilla but an independent, community-developed project.
Thunderbird has been re-added to the products steered by the Mozilla Foundation. https://blog.thunderbird.net/2023/11/the-untold-history-of-t...
I didn’t know that. Thanks for the update!
dubious, disingenuous claim.
Apart from the UI which is crap since their last major update. There are menu options everywhere, two ribbons on the top, a hamburger menu on the right and another on the left. For a long time you opened Thunderbird and it didn't default on the last message that you received but somewhere in the middle of the heap.
> Apart from the UI which is crap since their last major update
But when they updated the UI, they
- Added options to use to make it very close to the old layout
- Set those options for you if you had it customized like that in the previous version
Which is IMHO much better than how Mozilla handled the redesign - you can get the old style in a GitHub repo thanklessly maintained by one person[0], enable userchrome support in about:config (until they decide to take it away one day!), and enable compact mode (also gated behind about:config and called "Compact (not supported)". Oh, and remember to update the userchrome every few updates because they keep breaking it.
That's the difference between user-centric and not user-centric.
[0]: https://github.com/black7375/Firefox-UI-Fix
It's hard to be in charge of a project like this. You're criticized no matter what you do.
The old UI was criticized by some for being outdated, a mix of old and new styles, didn't fit well with new OS/app styles, etc. It was crap. So they update the UI and it's still crap... for other users. Damned if you do and damned if you don't.
the people still using a fat client are likely doing it because they don't like change
edit: I say this MYSELF as a thunderbird user!
I don't want an "new experience" every 9 months, and having to explain it to my parents
As a longterm thunderbird user I find this annoying. I appreciate it being maintained more actively again but I really liked the fact that the UI stayed stable for years. Changing things to make them "more modern" is just annoying. No one asked for this.
I was asking for it :/ (see: wishing vainly while project development appeared to be petering out)
Agreed, so tired of the endless treadmill of 'modernizing' UI
When you delete a message after that the old message remains selected and so if you hit delete again thinking the last time didn’t do the trick, you deleted another message and now go to deleted messages and try to find what you deleted.
The app has a phantom message even in empty folders that it keeps selected. Unread bubble and nothing else, an empty message. You can’t even delete it. Sometimes it persists between app restarts.
It shows unread count on a folder just because it feels like it.
It’s a long list.
Can I replace Outlook yet?
I don’t buy the “oh well, kinda sort of for like 60% of mail features and possibly a read only calendar in Two Weeks”
I switched away from Thunderbird to Outlook TWO FULL DECADES AGO, and in that time they have never once given me a possibility to switch back.
Like it or not, business runs on Office/Outlook.
You must be joking. The Supernova UI redesign is an unmitigated disaster. They unnecessarily butchered the look and feel of Thunderbird to the point where people are switching to forks.
Absolutely. New Thunderbird sucks, I was afraid this would happen when they resumed development. what’s the best alternative?
Give me the 1990s GUI back.
Clawsand SeaMonkey both have a more reasonable design. (SeaMonkey did a piss poor job picking screenshots for their webpage...)
https://www.claws-mail.org/screenshots.php
https://www.seamonkey-project.org/doc/screenshots
Well I like Supernova better, it made me switch to Thunderbird from em Client. (I always wanted to switch to Thunderbird, but until Supernova I never liked the UX enough to keep using it)
The previous UI made no sense. Having the email viewer on the bottom and the list on the top instead of side by side makes absolutely no sense when most emails are designed for viewing in portrait.
What are you talking about? Thunderbird has barely any progress in the last years. It's more busy with breaking and fixing things. Sure, there are reasons for it, but as a user, all I see is stalemate, while one addon after another is dying. Thunderbird Mobile is nice, and I hope Thunderbird Pro will be something good, but so far none of them are the big breakthroughs.
I ended up switching over to Betterbird. It's easier to setup and more stable.
Until it's not maintained, like most Thunderbird forks so far
Does it work with the OWL extension?
If you mean Exchange, then yes. It also supports OAuth2.
Where are the AI features in Firefox? Looking around right now the only one I see is right click tab -> Summarize page (NEW). I googled a bit and see they have some grouping of tabs feature I've never used/seen (or want). The only other maybe AI feature I remember seeing is the odd left hand bar that is there on fresh installs and I usually remove to declutter.
Are those the features this kill switch removes or was there a deeper issue here?
Firefox mentions the following ones:
"- Translations, which help you browse the web in your preferred language.
- Alt text in PDFs, which add accessibility descriptions to images in PDF pages.
- AI-enhanced tab grouping, which suggests related tabs and group names.
- Link previews, which show key points before you open a link.
- AI chatbot in the sidebar, which lets you use your chosen chatbot as you browse, including options like Anthropic Claude, ChatGPT, Microsoft Copilot, Google Gemini and Le Chat Mistral."
I wonder what sort of user testing made them decide that what Firefox users really need is a chat bot in the site bar. Isn't a chat bot in a tab good enough?
And calling translation "AI" seems like deceitful retroactive rebranding. Why is machine translation suddenly "AI" now? It was never branded as such before. Is "AI" here just used to mean machine learning?
The word artificial intelligence was coined in a 1955 research proposal [1] which listed seven aspects of the "artificial intelligence problem". Computers using languages was one of them. Another is "neuron nets", which would indeed encompass a large part of ML and at least Google Translate since circa 2016 [2].
This is also perfectly in line with how the word AI was used until circa 2022. The weird thing is this narrowing of AI to only mean transformer or diffusion based neural network approaches.
And many translation approaches would even fall under that, so not sure how narrow you perceive the term to be now. How do you even define AI to include everything OpenAI calls AI but not include modern translation approaches
1: https://www-formal.stanford.edu/jmc/history/dartmouth/dartmo...
2: https://web.archive.org/web/20180507195240/https://ai.google...
IMHO no. Every chatbot has so much wasted space, it really doesn't need to be full-width. Also, what's easier?
Option 1: Being on a tab, copying the URL of the tab, switching to the chatbot tab, pasting the URL and writing some instructions about what to do with that tab.
Option 2: Clicking on the "summarise page" button (whether from the sidebar or from right-click context menu), and having the browser pre-fill the prompt with the URL + the reader view version of the content on that page.
I saw someone else using this feature. It does more than just be a chat bot. You can direct it to automate tasks like go to a web page, search for stuff, etc. -- I asked it to go to pinterest and download the top ten images for "cyberpunk," and it succeeded. Nifty I suppose.
"AI" is a term which means a dozen things and has changed a dozen times. It's about as meaningful a signifier as "smart".
If I were to draw a line, I'd say AI is anything with a transformer model powering it.
As exhausted by 'AI' as I am, translation is one of the things neural networks (and especially transformers) have been constantly improving SOTA on.
They use a Neural Network engine to power it. That definitely counts as AI:
https://aclanthology.org/P18-4020/
I've run an AI translation tool under a debugger just to see what it did.
It tokenized your input, fed it into a model, then ran the model. Literally the same thing as any other local AI software. Except the model was for translation.
> I wonder what sort of user testing made them decide that what Firefox users really need is a chat bot in the site bar. Isn't a chat bot in a tab good enough?
If it wasn't because I find myself using the AI-sidebar all the time I would probably have shared your opinion.
I guess it's just quite convenient to have it separated from the "regular" tabs and their history.
The transformer architecture that powers large language models was designed by Google for the purposes of machine translation. As others have said, ML and AI have always been closely related if not synonymous.
Wouldn't be the first time. Google gave an option to turn off gemini in Gmail, and suddenly the inbox tabs they had for over a decade decided to disappear.
Firefox's translation specifically is from Project Bergamot: https://browser.mt/ It's a set of language models in the style that people currently call AI.
It is annoying. LLMs are not AI, and on-device translations are one of the few genuinely useful features Firefox has shipped in the past few years.
> AI chatbot in the sidebar, which lets you use your chosen chatbot as you browse, including options like Anthropic Claude, ChatGPT, Microsoft Copilot, Google Gemini and Le Chat Mistral."
When i saw this i expected something more... integrated, but when i tried it with a local LLM (using koboldcpp) after enabling the option to show localhost as an option (it is hidden by default for some reason) all it did was to local whatever webpage was running on the localhost URL (even though koboldcpp also provides an OpenAI compatible endpoint, which is what i expected Firefox to use to provide its own UI). It seems to have some sort of heuristic to find the input box where you type in queries and autofills that with the page text or parts of it (if you have it selected) and that's all.
I kinda expected it instead to use the API endpoint, have its own chat UI, provide MCP tools for accessing and manipulating the page's content, let you create reusable prompts, etc. The current solution feels like something you'd throw together in a weekend at most.
The chatbot providers don’t allow any cleaner integration unless you are using pay-per-request API rates.
They all seem like great features.
On paper yes. The problem is that they clutter the UI, they trigger at weird times and they turn out to be less useful that they may appear.
Then there's also people, like me, who just want the browser to browse the web. I don't want link preview (annoying feature), Firefox isn't my PDF viewer, I don't have that many tabs that I need to group them and I don't use AI chatbots.
So having a single button to disable all of these features is pretty great. I still want a Firefox Lite, that just does browsing and allows me to add the few extension I want to whatever feature I believe is missing.
I started clicking a 'next page' link before I'd actually finished reading something (so I kept holding the mouse buttown down), and a couple of seconds later Firefox popped up a 'link preview' box informing me that I was clicking on a link to a web forum. Wow, thanks, couldn't have figured that out myself. (It did not actually summarize the next page in any way.)
Whoa, I have been looking for AI-enhanced tab grouping for a while. This is actually pretty awesome.
I get some kind of AI summary when I do a long left click on a link. I'm sure I'm not the only one to accidentally trigger that one.
There's also a side window you can open that can connect to a chatbot. There's translation (on device). Also semantic history search.
Page translation is mentioned in TFA. It appears in the address bar on pages detected to be in a foreign language, and is also in the main hamburger menu.
Recently I also discovered about:translations. Dunno why it's hidden from about:about.
Wow, didn't know about this thanks.
The translation feature should be entirely offline and actually predates the AI everything push. I think it's a great feature.
More are coming, so it’s important to have a single clear toggle for this that respects user preferences.
YouTube is completely broken with this shitty spark thing. Also yesterday I was greeted with an entire new sidebar. It's comical
Ever see colored circles in your tabs?
No, I don't think I have?
Lol. At first I was thinking it was a AI kill switch on web pages (like Google overview...). I guess was being naive that they would do that, and also weird because there's barely any AI stuff on Firefox indeed...
I switched to LibreWolf, use uBlock Origin and have modified the web search so it uses noai.duckduckgo.com and appends -ai to every Google search query. I'm rather happy with this setup.
I wasn't aware of the "-ai" flags. Good to know.
Do you have a preferred way of blocking domains from appearing in search results? I think there's a limit of only five for some reason.
> Do you have a preferred way of blocking domains from appearing in search results?
Paying for Kagi which lets you promote/demote/block domains to your heart's content.
[flagged]
That's not a real thing, just a side effect: AI "quick answers" don't appear when you use search operators. In this case, you are using the minus operator to filter results with the word "AI", which might not be desirable.
I didn't know about ddg's noai subdomain, normal ddg has become unusable with all the ai shit they constantly shove down your throat.
Also, their ai makes copilot look good.
Good. I was fearing Firefox would also end up having too many AI-Features i do not want. But switching to Chromium-Browsers isnt an option anyways because of their Manifest V3 extension model. Restricting blockers? Whats next?
I don't know...at one point I got off Firefox because it was slow and I was never able to get back to it ever again. Maybe I should try now?
Do it. It's the only truly independent browser left.
It's not perfect, but it works, and unlike Chrome you can have full ad blocking with uBlock Origin.
> only truly independent browser
Only truly independent browser engine left. Firefox is entirely independent on google, but unlike its competitors this dependency is through direct cash payments.
Perhaps the Europeans will end up funding Firefox to get away from American corporate tech dependency. Firefox has tried valiantly to get away from Google money but it's difficult to fund browser development from small donations or without cross-subsidies from other business, like web advertising.
With Brave you can have ad blocking built into the browser itself and not have to depend on a third party developer.
And Brave's built-in blocker uses uBO's and other lists, including allowing trusted filters (which can inject high risk scripts into your sites) automatically. So yes, you are using 3rd party scripts already, and no, they can't monitor all of the commits from all those lists.
The first party developer in case of Brave is arguably worse than most 3rd party developers elsewhere.
I would argue that third party developer is more reliable and trustworthy than both Brave and Mozilla
Why do you actually ask? Switching browsers has got to be one of the easiest software to switch, right?
Is it? I have autofill/passwords, bookmarks, etc., and you can't easily jump between two different browsers as those will get out of sync.
Both browsers offer import/export functionality for all of those things. Spend one week trying Firefox and if you don't like it, switch back. The inconvenience here is so small it might as well not exist.
After I upgraded my laptop (Ryzen 2700 => 8845HS) it has felt as though Firefox is much closer to Chrome.
While I do not think that the gap narrowed when measured in CPU-cycles, it's just not very noticable when Firefox doesn't feel slow.
I went back to it 6 months ago once Chrome got rid of ublock. It’s pretty good but then some update a week or two ago blew away all my profiles.
The best time to plant a tree was 20 years ago.
Mozilla did plant the tree 20 years ago, then decided few years after to abandon it
They push millions of lines of code every year, thousands of patches and all kinds of measurable performance improvements so not sure what you're talking about.
Give it another go. You'll be surprised.
This just blocks AI features within Firefox.
The feature I would really want here is a switch that blocks AI summaries, overviews, etc. on any websites you browse.
That's the job of uBlock origin.
Eg here's a list
https://github.com/laylavish/uBlockOrigin-HUGE-AI-Blocklist#...
Unsurprisingly, ublock is still the best extension to do that, there are community driven lists that hide summaries and spam websites.
Much of the content online is already AI generated, and many websites have their code AI generated, so you'd soon be left with nothing but blank pages. :)
I'd rather blank pages than consume literal slop.
That's massively out of scope for a browser, that sounds like something for an extension.
[dead]
In addition to completely disabling AI, I found the following setting extremely convinent to disable in about:config. They clutter up my right-click on a link or on text selection.
Now if only I could get rid of "Print selection" and "Services" when right-clicking, too (on MacOS)Try modifying a webpage’s source to open with <html mozdisallowselectionprint>. If that works, you can make that universal using any page-mod plugin (though not necessarily the userstyle CSS-only ones). I’m mobile-only tonight so I can’t test myself, but derived from:
https://searchfox.org/firefox-main/source/layout/printing/ns...
Why would you disable "dom.text_fragments.enabled"? Why "privacy.query_stripping.strip_on_share.enabled"?
The latter, especially, seems helpful.
`dom.text_fragments.enabled` to remove "Copy link to highlight" on right-click. Yes, it can be useful, but I never use it and the very rare occasion of needing to use it when opening a page, I can just search it myself.
`privacy.query_stripping.strip_on_share.enabled` to remove "Copy clean link". I would rather it just did that clean link thing automatically, but I don't actually care about clean links -- it's just annoying having two "copy link" next to each other (especially with one which is greyed out 99% of the time!)
It strips query params used for tracking. Has it never bothered you when you click share on Instagram or YouTube and it inserts a unique-to-you "share ID" in the URL? I was burned by this once, now every time I copy a share URL from a social media platform I first paste it into my text editor and remove all of the privacy invasive tracking cruft.
As an aside I think it's only matter of time before this is done without query params and instead each share link is generated just for you.
You may be interested in the ClearURLs extension: https://addons.mozilla.org/en-US/firefox/addon/clearurls/
I'm also missing:
The last one removes the "Services" option when right-clicking an image or highlighted text.I wish we could also disable "Send via email" when right clicking a picture, I constantly misclick and userChrome.css does not work for the context menu on macOS since it's natively rendered...
> userChrome.css does not work for the context menu on macOS
Putting:
#context-sendimage { display:none!important; }
in that file works for me.
Does not work on macOS, it uses native context menus that can't be styled via userChrome.css.
print.enabled should disable all the print stuff.
Thanks! I didn't know that.
I think that's a good workaround, but I'll have to re-enable it when I actually need to print something.
Ironically, I bet that a significant majority of the users that turn on the AI kill switch — which must have some kind of phone-home telematics attached — will also be users who have disabled Firefox metrics collection and so will not have their opinion counted.
So, the most effective path here for y’all to be heard is not flipping the switch off yourself (do so anyways!) — anyone who cares at this stage has probably opted out of being counted already, after all — but instead to ensure that news of this switch spreads to absolutely as many non-tech people as possible. Don’t argue that they should run some script that shuts off their metrics and phone home and updates. Just convince them to shut off the AI and explain that this is why their browser got slow about a year ago! They’ll flip off the switch gleefully, their phone-home will count them, and y’all will have the strongest possible impact on the telematics graphs at Mozilla.
I already ran the disable process manually on the computers I have friends and family IT duties towards, so I’ll go back and do the AI switch to be sure it’s counted next week. Yes, this is a crap way to be heard. But making a mark on feature opt-out graphs is probably the only hope we have left to get their executive leadership to stop drowning the browser for its own good.
The other thing people can do is install Firefox and use it. An uptick in user share also serves as a metric to reinforce the move. Let's be honest, most people complaining are using chrome or some flavor.
But current Firefox users could probably temporarily turn on telemetry, activate the kill switch, and turn telemetry back off. Just make sure you wait long enough to ensure the information is sent
They should get the best metrics out of update server communications anyhow, 30% or more users getting update downloads than having AI enabled should be obvious.
I've read this three times now and still don't understand you. Do you think people don't update their software? Are you just counting every update as people happy about the kill switch?
There'll be so much noise in that signal it'll be almost useless. You can't differentiate it from anything else. For all anyone knows it happened because the word Firefox was in the top story on hacker news
If my choice is between a single blatant signal of hostility to AI that can’t be misunderstood, and hoping that a pro-AI company’s executives invested in voluntarily correlating two different sets of logs to prove itself wrong, then I’m taking door number one.
> will also be users who have disabled Firefox metrics collection and so will not have their opinion counted.
Gee. If only there was a way to collect users opinions on things. Welp.. guess we have to live with subtly spying on everything they do with our software.
Most people aren't vocal.
Most people who are vocal aren't representative of users.
Many vocal people aren't even users.
Don't get me wrong, I turn off telemetry, but you're acting like it's easy to get that information. You act like people don't scream when Firefox prompts people with surveys. You act like there isn't bias in survey takers.
If you just pretend everything is easy we'll just end up reinventing the same evils we're trying to fight today. Unfortunately most evils are created from good intentions. I hear there's an entire road paved that way
> but you're acting like it's easy to get that information.
It is, relatively speaking.
> You act like people don't scream when Firefox prompts people with surveys.
Surveys without proper response and adjustments aren't passing feedback, it's political theatre. People groan about surveys because it takes time and rarely shows results reflectant of the responses.
We know the system is broken. Hard to shame us into thinking we're the ones who broke it.
> You act like there isn't bias in survey takers.
You act like statisticians don't spend half their field accounting around bias.
>we'll just end up reinventing the same evils we're trying to fight today.
Let's have the old evils dealt with before worrying about creating new evils who happen to do the exact same thing as the "old evils" (spoilers: they are the same picture).
People who aren't vocal tend not to care very much. People aren't vocal tend to just accept whatever happens to them. So why not cater to the people who are vocal?
Mainly it's just developers not wanting to hear what people say.
Also, people lie. I don't trust what random users tell me, because years of tech support taught me that they're lying.
If we replaced telemetry with some sort of survey emails and phone calls, we'd get exactly another 500-thread discussion on HN about how "Mozilla is collecting emails to sell to the highest bidder!", "Mozilla is sending us spam!" and whatnot.
I think at this point they know all these opinions pretty well, but they simply don't care or see better growth options by targeting users who don't belong to that particular bubble. They see OpenAI approaching the same active user counts as Facebook and they want a slice of that pie. And the majority of that pie is non-techies.
Yup. When people feel like their voice won't matter, they will simply shut off any channels "listening" to them. You need to build up trust first.
As a complete contrast, there's a lot more enthusiasm over Valve surveys.
Arguing that telemetry is wrong doesn’t seem to have stopped Firefox from using it. If your fight against telemetry is a higher priority than your fight against AI in the browser, good luck and more power to you! I’m just making sure no one makes that decision by accident.
This isn’t about a fight between two things. This is about a company that completely refuses to listen to their users and then some random guy on the Internet who says you should deal with their telemetry to somehow fix the problem that in no way relates to the actual problem and no way has evidence that they would take steps to fix the problem either. It’s very prosocial to think that Firefox actually makes changes based on user feedback, but almost none of these features have users been clamoring for, this is totally executive bullshit pushed into the product.
For anyone motivated to do this the number of installs will be vanishingly small.
They'll just push an update that prompts users to re-enable AI until their metrics have reached the relevant manager's targets.
Telemetry was never about user preferences but all about justifying what you were going to do anyway.
The best thing about Firefox telemetry is that it can't be easily disabled. There are many setting that control it. Including an external scheduled task that can't be disabled using Firefox itself. And even if you delete the task it will come back after update.
You wouldn’t know it from how often scripts claiming to disable telemetry hit the front page.
Great, let's see how it works out.
Firefox for Android has been killing it for me with the latest ux updates, I didn't expect major improvements there and was pleasantly surprised.
I don't see the appeal, it takes more "clicks" to do many actions and I had to disable the ridiculous new oversized "rectangle tab preview block" (whatever it's called).
yes exactly, their design was already better than chrome and condensed but now we have these outdated round and padding heavy toy controls again, just why?
I actually liked the previous UI much more, the new one looks like a baby toy and uses more space because of the control padding. completely unnecessary.
This UI is great but do you get this horrible thing where sometimes the browser is shows a white screen and you have to force stop the app? Happens all the time on the latest version for my Pixel 9a. And did on my Pixel 7 too before. It's really horrible and I can't pin down any rhythm or reason other than loosely seeming to happen more often when I'm in battery saver mode.
I don't have that on fenec and ironfox on my pixel 6a running graphene.
What I have always had with any firefox based browser on android is erratic behavior in text field. Most of the time it works well but sometimes on some commenting systems my input is duplicated/multiplicated/garbled, trying to select where you want to insert words in the middle of the sentence sometimes becomes impossible, always resetting to the beginning of the text field, etc.
On some websites it only rarely happens, on some it is much more regular. Never understod why but when I want to edit a comment I have to resort on a regular basis to copy the full comment to a note app, edit the text, and replace it in its entirety in the text field.
Super annoying but still less annoying than using a chrome based browser with no way to remove ads and have a bit of privacy control.
Happens to me all the time in PWAs. I get it roughly once in two days, sometimes a few times a day.
I have also noticed that it seems pretty common with PWAs so I more or less stopped using them (in the sense of installing them).
Are you using some weird extension ? Never had this on my pixel 6.
Only extension is ublock origin and most of the settings are default.
Not yet for me on a pixel 6a
Never
not happens on my Poco
Never. Using a Snapdragon phone. Sounds like an another isue with the Tensor SoC...
for me, the killer feature is that I can use uBlock Origin.
This and Violentmonkey which syncs with my desktop Firefox so I have all the same scripts. Wish it would also sync the script data though.
Which UX improvements in particular?
Agreed, was pleasantly surprised and didn't know how much I didn't like the old one until seeing the new one by comparison.
How do you make closing tabs from taking literally seconds.
Firefox does what some people want, people complain. Firefox does what other people want, people complain. Firefox does what both people want, people complain.
I'm sorry, but we'll never get corporations to do what we want if we don't throw them the smallest bone when we get our way. You need positive reinforcement too, not just negative. If it's all negative they just stop caring and you get companies lot Google who just don't give a shit anymore.
And yes, there are some AI features I like and I want in the browser. I get a lot of utility out of translation as well as semantic search of my history. I don't want agents in my browser but get, Firefox is giving us choices.
Look, no one needs to like Firefox, but let's also be honest, it's the best we got right now. Google, Apple, and Microsoft are shoving agents down our throats and putting us in walled gardens that are getting harder and harder to break from. I don't care what flavor of chromium you use, Google is still using it to control the way the web works. Everyone loves to say how chromium is has greater coverage of standards but never takes a second to question who sets those standards.
I'm sorry guys, that's the state of things now. You can't fight Google by switching to chromium. It's still their vehicle to eat the internet. Our choices right now are Safari, Firefox, and maybe ladybird. It's slim pickings and nothing is close to perfect. At this point it doesn't even matter if Mozilla is evil, because at least they're the enemy of our enemy. Google is keeping them on life support to avoid monopoly claims but how long will they need that?
So what, we're just going to hand the keys of the kingdom to the guys selling artisian turd sandwiches because what, there isn't enough mayo on your ham sandwich? Because you don't like ham?
We got a win. Celebrate. Take the break from being cynical. There's bigger battles to fight and there'll be more tomorrow. Take the night off and don't be a sore winner
Nowadays, I avoid certain topics. For instance any post about macOS becomes one giant complaint.
Thanks for staying positive. I like Firefox, I think it's a very nice holdout against adware.
> So what, we're just going to hand the keys of the kingdom to the guys selling artisian turd sandwiches because what, there isn't enough mayo on your ham sandwich? Because you don't like ham?
Firefox is the artisan turd sandwich. They are burning dev time on features barely anyone asked, while bleeding market share for last decade
This is a narrative that got hallucinated into existence by comment sections. Most of the effort goes towards the core browser and always has, side bets have found their ways into headlines and comment sections from users overestimating the resources necessary to build them out while underestimating the scale of regular work like security patches and performance improvements.
> They are burning dev time on features barely anyone asked
Got a source for that? HackerNews is *not* representative of the average browser user.
The average browser user isn't using Firefox and won't be any time soon.
Okay, so in your version who is Google? And chromium?
The reason people complain is because Mozilla claims it's better and more pure than everybody else.
> people complain is because Mozilla claims it's better
Well they are far from perfect but experience has shown that everybody else is worse.
Well "everybody else" is Google. I'm not sure that bar is even high enough to trip over.
Google doesn't sell my data. Mozilla does.
Corpos are not entitled to bones.
Who has expressed a desire for Firefox to become "an AI browser"?
Because that's the source of the complaints. I don't want to use an "AI browser", kill switch or not. If this "AI browser" dies because of their mission to destroy community goodwill, good. I'm sick of giving the benefit of the doubt every time they royally fuck up. This situation where they're the steward of the only non-Google browser is not tenable, something needs to change.
Is it so hard to understand that the features person A wants are not necessarily the same as person B yet a company/foundation would be interested in providing a complete featureset for both A and B?
I don't need themes nor having my url bar serving search. I am not interested in an AI agent in my browser yet I welcome traduction features, do I have to shit on every company developping a software that has some features because I don't want them?
I am much more pragmatic: are these features easy to ignore/disable, do they largely increase the resources needed (disk, memory, cpu) even when not used, do these introduce bloat, etc.
I wasn't interested in pocket, I was just using a combination of firefox forks or disabling it on the devices I was using. That is the whole point of open source software.
But honestly it just feels like you didn't even read my comment. I'm sorry that it's a lot, but I'm petty sure people can handle 10-30 seconds of reading. I even said it doesn't matter if Mozilla is evil. How do you turn that into me giving them the benefit of the doubt. I'm literally just arguing that there's slim pickings and to not help our bigger enemy to kill their enemies. It doesn't matter if their enemies ate evil, you're just helping the bigger evil get bigger and consolidate power. I'm saying "there's more important problems right now, not be fucking dumb and get distracted or before you know it you'll lose your head"
Letting Mozilla torpedo the only non-Google browser is also a bad way to fight Google. It's looking bleak. The only hope is that Mozilla dies and someone more serious picks up the mantle. Not sure who that would be though.
If I smash your window every day but one day after being asked to stop again and again I decide to just leave poop on your front porch instead, should you not complain about the poop?
Malicious compliance is no compliance, it is still malicious.
Meanwhile I still do not have the option to change/create the profile from menu. I kow about about:profiles but I dont have this: https://blog.mozilla.org/en/firefox/profile-management/
It's finally there but you have to enable it in `about:config`. Search for `browser.profiles.enabled`.
You can do this in Vivaldi if you don't want to use Chrome.
Probably not enough for AI haters, we need separate executables with kill switch already triggered!
That would be better, yeah. Ideally some #ifdefs that I can configure to remove all the slop from the binary itself ;)
Really all this bloat should just be extensions instead of part of the base browser code base.
Firefox already captured the developer audience, and it wasn't enough. So they pivoted to whatever they thought of to increase mainstream adoption. That is really alienating that developer audience though.
Firefox had mainstream adoption in the Firefox 2 era, before Chrome launched. What drew people to Chrome was that it was fast, efficient, and did I mention fast? Unfortunately all browsers today seem to have forgotten how important that is.
The whole purpose of Chrome is to have google search by default in people's browsers. Once they have a large market share nothing else really matters. If people truly cared about speed Google would have changed it already.
What drew people to Chrome is that Google leveraged their search monopoly to push it, and paid other software to sneak it into installers.
Am I the only one who absolutely adores the local based translation feature? I use it regularly and its so good.
Yes, I really like it. I wish they let extensions access it directly but it seems they don't (yet, at least). I'm also a bit surprised more hasn't been done to distribute it separately as a command line tool as I think something like that would be very helpful. I have seen a couple of attempts to put a CLI wrapper around it but they are third-party and seem to not be well maintained or documented.
Some one up thread pointed out about:translations is where it seems to be hidden.
Also a fan of this feature. It's actually been around awhile but I think the Asian languages are a more recent addition.
I wish there were some updates about PWA support. Haven’t heard about progress on this since last August. Is it still in beta and only available on Windows?
At this point, the industry wide decision to have AI as opt-out is suspicious. Why not have an AI opt-in default? It would certainly save so much trouble from the Luddites. On the other hand, an AI opt-in default would be not using any technology at all (where opting in would be turning on your computer or device) under a dictionary definition of artificial intelligence.
I'm torn on whether to see this "AI Kill switch" as a win on respecting the users, or something to keep us distractewd while they ship through "Trusted Types" API that sounds like further restriction of end-user computing freedoms.
I would absolutely love to hear your reasoning that leads to type systems being considered a "restriction of end-user computing freedoms".
For those that don't know what trusted types are: Simply put, it splits the string type in to unsanitised_string_from_user and safe_escaped_string where unsafe strings can not be used in function parameters that only take a safe string That's heavily simplifying of course, but it's the basic idea.
Skimming the API docs on MDN, it makes sure the site vendor gets to run filtering code over anything you'd want to inject via e.g. user script or console, securing it with CSP. I expect this to make user scripts work as well as they do on Chrome now. If there's a workaround, I'd love to hear about it.
Oh just great. The web is bad enough already, I think I'll have to go live in the woods if userscripts get kneecapped
Worst case you just run your userscript before any policies are created, but in most cases it's not going to impact userscripts.
You may be thinking of the much-hated "Trusted Computing" initiative. "Trusted" here means that the JavaScript dev picks a sanitizing library they trust, not that Mozilla decides what software is trustworthy.
Nah, my issue isn't with users vs. Mozilla, but users vs. "JavaScript dev", specifically the difference of opinion on who should have final say on what gets executed and what doesn't.
Aren't those just overengineered sanitizers?
Question is, can you sidestep or disable them in user scripts or in developer tools, without disabling CSP entirely or doing something even more invasive (and generally precluding use of that browser instance for browsing)?
We made sure to exclude WebExtensions code from web pages's Trusted Types restrictions enforcement. (Bugs can happen of course)
Step 1: Launch AI features Step 2: Launch AI features kill switch Step 3: ???? Step 4: Profit?
Search licensing that Mozilla currently depends on might become eclipsed by AI so it's important to ahead of the game whatever it might end up being.
Yeah, Mozilla made us do an additional step here.
Before, we did not need to disable AI stuff. Now Mozilla forced us (that is those of us who don't like or use AI) into an extra step. Guess the only thing worse is being given no choice at all though.
Did they also fire the CEO who wanted to make Firefox into AI browser?
It is such a shame where Firefox is and is headed these days. It could have been a browser for the people, instead it's an advertising machine, similar to Chrome.
Icecat/Iceweasel are sane alternatives.
Unofficial binaries can be found:
https://icecatbrowser.org/download.html
Official source: https://git.savannah.gnu.org/cgit/gnuzilla.git
I would prefer it, if the basics, like being able to use your Webcam would work. That's broken for a while now on Fedora/Linux
Works for me on CachyOS and previously Pop-OS.
Used my webcam just fine multiple times today on Fedora Silverblue.
I'm using the Standard Mutable Version. But maybe I should try replacing the shipped Firefox, with the Flatpak one.
Right now I'm switching to Chromium for Video Calls, which I hate
Can I turn off Firefox's AI?
I keep bumping that button on mobile and it's killing my experience.
last time when I updated Firefox, the package manager began building ONNX Runtime from source, which my "minuscule" 16GB of RAM couldn't handle. I want that during install time, as I don't like the idea of rebuilding ONNX every time Firefox updates, period.
That is an issue with your OS. Your OS vendor should be precompiling everything for you.
Gentoo ?
no, NixOS Unstable. normally, they precompile such stuff, but ONNX decided that it wants to link against my ROCm instance specifically. looks like soon we will have to resort to dirty workarounds akin to those in Blender (HIP, CUDA support)
Will it render &em; as &en; ?
Why would it to that? -, –, and — are all different things.
An AI kill switch is my most-wanted feature. AI is equally terrible and pervasive.
[flagged]
It's not all bad is it? On device translation of websites for example is much better than the alternatives.
Until very recently, on device translation was not marketed as AI.
Yet translation was the main application for applied language machine learning.
On device or not, it's a transformer model, so some view it as tainted.
I don’t want nor need on device translation enabled by default. I’ve gone without it for the three decades in which browsers have been around. I’m sure it’s brilliantly useful for some people. A one time ‘would you like to enable AI on startup’ for at least years with user profiles that are significantly old would at least be a show of good faith.
It literally breaks YouTube
It's more like if eating guano became a fad, and restaurants started offering guano dishes, and this one said "OK we'll still have guano on the menu because it is popular with some patrons, but we'll have a separate menu with no guano options at all for those who want to be sure they're not getting any."
You're analogy works but you can't forget that there other restaurants. That the other restaurant not only aren't making promises to not defecate in your for but they're actively advertising how much shit they can shove in a sandwich. Even the bread is made of shit!
So thank fucking god. At least there's one place where I don't have to eat shit. The bar is so fucking low it doesn't matter if they spit in it or you find the chef's ball hairs, at least it isn't shit.
"The 'AI Kill Switch' in Firefox 148 is not just a feature; it's an architectural statement on Agency. Most platforms today operate on a 'Seed-to-Harvest' model where the user is merely the soil for data collection. By allowing a hard-disable of AI functionalities, we are seeing the emergence of a 'Watchman' architecture—one that prioritizes the user's right to define 'Value' beyond algorithmic efficiency. We need more systems that respect the 'Seed' (the origin of choice) rather than just maximizing the output."
I love how AI's most-requested feature is always a way to completely disable it.
Maybe it's just me but why is it on by default? Why is this shit not off in the first place? Why can't Firefox just be a browser with great html, css, js rendering and then have a bunch of toggles for extra crap that people want? Do they actually have metrics that show "When we enable this crap by default we make an extra $X Million per year"? I'd put money on that being untrue. I bet it's like the data-driven ad spend - I've yet to see anything that proves that hoovering up bajillions of data points on each person moves the needle on spending beyond just showing a context-relevant ad, e.g. An ad for fishing gear on a blog post about fishing gear!
Honestly, I feel more and more every day like old-man-shaking-fist-at-clouds! Can we not just have something that works without spying, without engagement-driven shit switched on all the fucking time?
I think of the Simpsons Mr Brown meme where he's asking "Is it me that's wrong?".
I can't be the only person that thinks this way!
Because the vast majority of Firefox users aren't contrarians or diehard anti AI crusaders. These features are helpful to people. So they're on and available for people to use, and since it's a foss customisable browser, users with strong opinions can change it to their liking.
I don't think the AI features are used for advertising data harvesting but I'm happy to be educated if you have a source saying otherwise.
>Maybe it's just me but why is it on by default? Why is this shit not off in the first place?
Because there would be millions of emails to support asking "why can't Firefox translate a page the same way chrome can?" from people that couldn't find the AI opt-in switch.
> Can we not just have something that works without spying, without engagement-driven shit switched on all the fucking time?
Absolutely not! Making someone opt-in (the horror!) would result in too much confusion and support tickets! We can't ignore those tickets and tell people to figure it out, we have to cave and turn all the bullshit on by default.
I think a lot of our problems can be explained by: corporations lacking a spine and people unwilling to learn about the computer they use everyday. It's basically unacceptable to not know how to change a tire if you own a car but clicking through some browser settings is too hard, I guess.
Expecting everyone to learn a lot about Firefox to turn on features they want is a lot more of a burden on the users than expecting opinionated power users to remove features they don't want.
Firefox translation is entirely offline, so is the LLM powered tab groups. Nobody is making money off Firefox Translations.
Sounds like LibreWolf would suit you better: https://librewolf.net/
If I wanted a browser with AI, I would have used Chrome or Edge
(This post is directed to all software that shoves features like this in my face, and especially Microsoft more than Firefox.)
My problem with all software that shoves these AI features in my face, is that I don’t use features under duress.
If you interrupt what I’m doing to push me to use a feature, I won’t use it. If you’re a web designer and you block the page to tell me to sign up for an account, I close the tab and vow to never create an account. If you stop what I’m doing to ask me to rate your app, I’m going to give it 1 star. Et cetera.
Now I’ll be the first to admit this is childish… it’s a flaw in my character. When I feel pushed, I push back, and software pushing me makes me irrationally angry for reasons I can’t quite articulate. In some ways I wish I wasn’t like this. But I can’t be alone. I’m certain there is a non-negligible number of people like me, and when a browser immediately shoves AI features in my face on first launch, well, the first thing I’m going to do is disable them.
The especially tragic part is that I personally find LLMs useful! And I’m at the point where I sorta want to install a Firefox extension for ChatGPT now. But the actual browser AI features were pushed on me in a way that made me feel violated, so I can’t use them on principle. Maybe in a few years I guess.
If instead these companies would just dial it back several notches, I would have had the curiosity to try these features out myself, and I’d likely be using them by now. But the way they’ve tried so hard to force them on me has destroyed my trust and now, not only am I not using whatever feature they promote, I hate their product more than I otherwise would.
Firefox isn’t actually that bad here, and now that there’s a simple kill switch, I may actually try their chatbot sidebar thing. But for companies like MS, I will never, ever, ever use any of their AI features for the reasons above. (I’ve literally uninstalled Windows now, it’s gotten so bad.)
How is a company supposed to accommodate your personality while also showing other user new features that Firefox just added to the browser? If you're a contrarian by nature Firefox can't help you nor change their strategy to suit you, because the majority of users aren't like you.
You never, ever need to interrupt the user or stop them from what they’re doing in order to show them some new feature.
Here’s what you do:
- Make a “what’s new” section in settings.
- Put a link in there that takes me to a webpage where I can see what’s new.
- That’s it.
Instead what always seems to happen is that I’m in the middle of trying to do something with the software and in order to do that I must close whatever popup you’re shoving in my face to tell me about the new feature. I don’t have time to read it now because I opened the software with an intention to finish a task and I don’t have time to read it now. And then later when I finally do have the time to look at your new feature? Nobody bothers actually putting that information anywhere persistent, so I guess I’m out of luck even if I care about the feature.
Updating to a new iOS version is a perfect example of this. Say there are a dozen apps that have a new feature popup on first launch when you update iOS. Imagine a typical day waking up and trying to use my phone. I have to drive somewhere so I try to put an address into Maps and have to immediately fend off the “what’s new in maps” dialog so I can type the address I need to go to. Then I want to put on the song my kid is yelling at me to play and have to fend off the “what’s new in Music” popup. Later I’m trying to respond to an important text and have to fend off “What’s new in messages”, etc etc etc.
That first day using iOS is an absolute nightmare because of this.
Now imagine the alternative: a simple badge icon in the settings app, and I tap it and see a link to “what’s new in iOS”, and guess what, it can be a fucking webpage! I can bookmark it! I can add it to my reading list and see it later! Hell, I can even share it with my friends!
But no, instead apps insist on trying to increase “engagement” with their new feature, because some PM’s promo packet wants to include “this many users used my new feature” and the only way they can think to do this is to (1) stop the user from accomplishing their task until they tap the cutesy “Got It!” button, and (2) don’t bother with persisting it anywhere, because the idea that the user doesn’t have time to check out the feature now is so foreign to these sociopaths it never even crossed their mind.
Tragically, capitalism drives corporate behavior.
Whatever works for large numbers is what will happen.
But overall, you and I (and many) will try to push back and insist on consent.
The sign-up form with an unchecked "sign me up for your newsletter" option.
The first-run experience with a question... "do you want us to notify you of new features?"
But this is not the norm, and even if good actors get rewarded by a few childish customers, bad actors seem to get rewarded much more by a massive infusion of funds.
Why wasn't this there from the get go? Many people dislike the AI spam; I do too. I use chrome-based browsers usually (I also hate how dependent I have become on Google; default firefox refuses to play audio on my linux system as they claim we need pulseaudio, chrome instead makes no such assumption and audio plays just fine, so one can go and figure out why mozilla acts worse than Google here - all the google-bribe money killed its THINKING ability), so when I do, I use a few extensions such as "disable AI overview" or similar. It is annoying that we have to invest time in order to uncripple the world wide web. Browser vendors should be much more responsible, from the get go. But they all want to jump on the hype train, to milk out more money. Greed is the driving theme nowadays. (They could offer AI based on people who want or need that, rather than cram it down onto everyone.)
> Why wasn't this there from the get go?
Even better, why was the AI feature ever added in the first place?
Quite a few of the LLM features actually add value for a certain group of users. Automated image descriptions for the visually impaired, automatic translation, ... Running those on local models is a net benefit for quite a few people, but they get a bad rep because they're "AI" and the current trend of shoving AI everyplace and with no means of escape means that AI in general has a - well deserved - bad reputation.
Because a browser needs users, and some people like AI features. Firefox can't win the battle, or even survive, on an AI hating, nerdy user base.
Is there disable auto-update setting in GUI? Last time i looked there was none and i had to create some settings.json file for that.
That's why I use Helium now.
yet another chromium clone iirc.
It's basically ungoogled-chromium with manifest v2 support. Chromium is just technically superior than Firefox. It's a simple fact. The problem is the telemetry and AI features they added in it, which Helium or ungoogled-chromium doesn't have.
>It's a simple fact
I'd love to hear why you say this.
In case you're looking for the opposite, not an AI "kill switch", but a propaganda kill switch using AI to fight disinformatiom, try BiasScanner:
BiasScanner - Firefox Plug-In https://addons.mozilla.org/en-CA/firefox/addon/bias-scanner/
Paper: https://arxiv.org/abs/2407.10829
Now I need a switch for my smartphone and my computer too.
Another opt out anti feature. Luckily better forks like Mullvad Browser and LibreWolf exist that actually deliver what Firefox promises.
Firefox is the only holdout against the ad companies, and I'm counting Microsoft amongst those. It's a very good browser, independent with its own renderer, with decent ad blocking and decent performance.
It continually amazes me how people use a Google product on their desktop, as if they don't send enough data to an ad company. Actually, I'm not sure why I type this, any rational arguments are definitely not winning them over.
[dead]
In theory you are not incorrect, but Google bribes Firefox and Google makes most money via ads. Mozilla gave up on firefox a long time ago.
> It continually amazes me how people use a Google product on their desktop, as if they don't send enough data to an ad company.
I'd love to have alternatives, but which ones are there? Firefox is not an alternative; audio does not work for me as I am pulseaudio free here. On chrome-based browsers audio works fine, out of the box, so it is not my system that is at fault; it is mozilla that is at fault. I also reported this, the lazy firefox dev said all Linux users use pulseuaudio these days. Well ...
I could recompile it but compiling firefox is a pain in the ...:
https://www.linuxfromscratch.org/blfs/view/svn/xsoft/firefox...
I am not going to use a build system that is +20 years old and only exists because Mozilla is too lazy to switch to cmake or meson/ninja as primary build tool.
> Actually, I'm not sure why I type this, any rational arguments are definitely not winning them over.
Well I gave one rational argument: can't play audio on my linux box if I use firefox (by default that is). I can give many more reasons too. You seem to make the point that Google is worse, so we should also use a bad product (firefox). I think we really need better browsers in general. Firefox simply isn't one and that is Mozilla's fault. There is a reason why it went into decline. Mozilla gave up the fight - the ad-money made it weak.
> Firefox is not an alternative; audio does not work for me. I could recompile it but compiling firefox is a pain in the ...
Obviously I don't have any data backing me up here, but I'm going to guess that that isn't the main reason why so many people choose Chrome over Firefox.
Firefox has been my main browser lately, and in my experience it covers pretty much every latest spec: no issues with Web Audio, WebGL (as well as WebGPU, I think), CSS features, etc. There are some select cases where Chrome has deployed something and Firefox is lagging (Background Fetch, for example) but that affects me more as a developer than a user. I cannot remember a single time when I opened something and it didn’t work in Firefox.
You made the decision to "pulseaudio free" your system, why do you expect others to fix issues arising from that decision of yours for you?
> I could recompile it but compiling firefox is a pain in the ...
Would second this. Mach uses Python, and the dependencies they use are a pain whenever no pre-built wheels are available. Especially so when you see that an "optional" Mach dependency for build system telemetry is what busting the configuration (not build) stage...
Do you mean you disable pipewire-pulse? Why? Or does audio not work for you with pipewire-pulse? I've never had issues with firefox and pipewire-pulse on my system.
I daily drive FF in desktop and Android but Brave has doubled in users the last few years, and my mildly tech-conscious acquaintances have settled on it after Manifest v3, while FF has been flat. That has been the greatest vote of no confidence against it ever.
[flagged]
You're being down-voted because it's a low-effort comment which comes with a large burden of proof that you've not included. Specifically:
> mozilla is basically a google subsidiary
"Everyone" knows that Mozilla has a heavy financial reliance on Google. So are you bringing this up to suggest that Mozilla also consistently acts to benefit Google and its ad network? If so, where's the proof? If not, what's the point you're making?
> firefox telemetry is almost comparable to chrome
Comparable to Chrome what? Telemetry? Something else? What is Firefox using that data for? In the service of or against users? What's the point you're trying to make? If you're making assertions, where's the proof?
You're making a lot of imprecise comments, most of interpretations of which carry a large burden of proof, and then complaining that people are just down-voting and moving on.
That's a pretty big aside
You’re mixing up funding with control.
Mozilla Corporation takes money from Google for search placement. That doesn’t turn it into a subsidiary. Google doesn’t own it, doesn’t run its roadmap, and doesn’t ship its code. Mozilla negotiates search deals the same way Apple does for Safari. Revenue deal ≠ corporate control.
On telemetry: you’re overstating it. Firefox ships with telemetry on, but it documents what it collects, lets users turn it off, and exposes most of it in about:config. Google Chrome ties into a much broader account system, sync stack, and ad network. Chrome doesn’t operate in isolation; it plugs straight into Google’s data ecosystem. Firefox doesn’t own an ad network to feed.
“Almost comparable” needs evidence. Comparable how? Volume? Type? Identifiability? Retention? Without specifics, the claim collapses into vibes.
The bigger difference sits lower in the stack: engine independence. Firefox runs on Gecko. Chrome runs on Blink. If you care about web monoculture, that matters more than marginal telemetry deltas. When one engine dominates, web standards start drifting toward what that engine implements. We watched that happen in the IE6 era.
As for uBlock Origin: yes, it’s a major reason people choose Firefox. But browser architecture shapes how long powerful content blockers survive. Chrome’s extension model changes (Manifest V3) restrict what blockers can do. Firefox kept the older, more capable API. That choice signals priorities.
If your argument reduces to “both collect some data, so it doesn’t matter,” you flatten meaningful differences. The question isn’t purity. The question asks who controls the engine, who sets extension policy, and who benefits from surveillance at scale.
If you think those differences don’t matter, make that case directly. But don’t blur structural distinctions into “basically the same.” They’re not.
That is not a mix-up though. Mozilla became dependent on the Google money - everyone sees this.
> Mozilla negotiates search deals the same way Apple does for Safari. Revenue deal ≠ corporate control.
Google search revenue represents about 75% of Mozilla's total revenue.
Google search revenue represents about 4% of Apple's total revenue.
If you think those differences don’t matter, make that case directly. But don’t blur financial distinctions into “basically the same.” They’re not.
[flagged]
[flagged]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
Thing is, there's a large (or at least certainly vocal) contingent of users (and mostly techies, to boot) that view "AI" as the Devil, and transformer models as the original sin, and they want to refuse to partake, wholesale.
This feature seems to be a nod to people with this worldview.
EDIT: See e.g. https://news.ycombinator.com/item?id=47133786 liking AI features to defecating on your food. It's not a technical objection, it's a principled one.
And that's in turn because product managers keep calling everything "AI" and shoving the bad kind in every feature they possibly can.
What is the bad kind that Firefox is shoving in?
Do they have any good kind?
What's the ratio?
I figure, hey, at least Mozilla listened and provided the opt-out. It could be worse. I also happen to be in the "food defecating analogy" camp, and I can give the developer an unenthusiastic thumbs up for at least listening to the peanut gallery this time.
Ideally they wouldn't make the product bad, with a badness opt-out, in the first place, but everyone in Silicon Valley's got to feed the AI monkey. So I guess this is the best we can expect.
FWIW, I may be in the other camp but I strongly respect them for providing this feature. It perhaps wouldn't be necessary if the pro-AI push wasn't so ham-fisted and utterly disrespectful of users for the past years.
(Also I didn't realize how bad this push got until I visited California recently, and saw every other billboard - that's physical ad over a road - pushing some unqualified form of AI magic on me).
More like excuse to add more of the features. "See, we gave you kill switch"
Exactly, what Mozilla is doing is malicious compliance.
I use AI for coding, I don't want it on my browser. Actually I don't want it on Firefox. Let Brave be the AI browser. They could have used AI themselves to actually make their pathetic new split tab view usable.
From TFA:
> For those who wish to maintain some AI functionalities, a selective blocking option is available, enabling users to retain useful features like on-device translations while avoiding cloud-based services.
Sure, but then the people complaining would need to recognize that that's not happening. All the AI features are local models. The only thing not local is that side window you can open that can connect to a chatbot. It's also easy to disable.
Honestly, I think people just like complaining. I think they like complaining about Firefox even more. There's plenty to complain about, but aren't there bigger fish to fry right now? Seems like complaining about some minnows while we're being circled by a bunch of great whites
I think a lot of people are simply sick of seeing "HEY NEW AI FEATURE LOOK AT ME" popups everywhere. Shove something into people's faces often enough and it becomes like a red rag to a bull. That's probably not the only thing, but it's one big motivation why some people want it.
Disabling the local spell checker when one disables AI is just malicious compliance.
People vehemently asked for a kill switch that does exactly that - kill off all AI-related features. I quite like the local LLM translations etc., but jedem Tierchen sein Plaisierchen, as they say over here.
Seems like willful disregard of the fact that "AI" has been now pushed as synonymous with chatbots, "AI summaries", LLM text generation when convenient, but then at criticism, it goes back to "oh but machine translators are AI too".
https://news.ycombinator.com/item?id=47077431, by dang, 4 days ago:
>"But I use it to help my English!!!! Who cares if it's AI if the comment is good??????????"https://news.ycombinator.com/item?id=46747998, by dang, 1 month ago:
If you don't flag this shit when you see it, HN is fucked. The commons is fucked. And you all keep upvoting and replying to comments from accounts that start posting 30 comments a day after 2 years of silence that are all one tightly packed paragraph of pablum with the same structure and tells that are harder to fix than replacing the em dash with a double hyphen.---
P.S. The actual comment is (surprise!) completely wrong. Visit Settings -> AI Controls and you will see a granular set of feature switches under the master kill switch. Each has a clear title and description and is independent.
Structure-wise, that comment is not great. Why do you think it’s AI — simply because of a dash of flavor?
Not to mention that an LLM would presumably use an actual dash, not two hyphen-minuses.
Click on the profile.
Read every comment made since the account started posting again.
Tell me what you think about those comments.
I don't think AI features in a browser are bad, and I think people who tut-tut it are overboard.
However, I think data control is critical and any kind of implicit cloud service such as transmission to remote AI servers should be toggle-able clearly, just like search autocomplete can be done.
[flagged]
Be nicer. You could’ve sufficed with:
> It's good that they've added an option to disable them for those who don’t want to use or see them.
[flagged]
We've banned this account. You can't abuse other community members like this, no matter how right you are or feel you are.
We detached this subthread from https://news.ycombinator.com/item?id=47133763 and marked it off topic.
I was being sincere, my friend. I genuinely envy that worldview. I long for it. I wish above nearly all else I could reset. I was being authentic and vulnerable. Why does that infuriate you?
Read your statement again and tell me who the pessimist is, and who is most in need of a break from the internet.
The legal structures that mandate what power google actually has over mozilla still presumably exist though. Pretending that we are in full blown dictatorship is, in my view, still cynical.
Though of course there’s no telling how far we will eventually go in a trumpworld.
> Pretending that we are in full blown dictatorship is, in my view, still cynical.
Could you please point to what I said that implies I'm pretending a "full blown dictatorship?" I apologize if that's somehow what I indicated. It certainly wasn't my intent.
You envy the worldview in which people back their opinions with actual arguments?
I envy the worldview it's possible someone can take money from another and still maintain independence.
What's up with the straw men?
i intended to ask what the difference was between two browsers that are both beholden to a company whose express goal is to suck up personal data. so far ive gotten vitriol, AI, and downvotes. my actual question remains unanswered. if you'd like to answer the question that would be cool! but yeah if you dont want to answer , it'd probably be easier to say nothing than to tell me to die alone
Wake me when they add webserial.
What an absolute dystopia when disabling AI is considered a "feature". Obviously the _enabling_ of it should be the "feature" if anything, and considering popular opinion there is absolutely no reason to enable it by default.
> considering popular opinion there is absolutely no reason to enable it by default
You need to update your priors. The popular opinion is there is a reason to enable it by default. ChatGPT is the #5 most popular website in the world, more popular than Wikipedia, Reddit or Twitter. The vast majority of users want to use AI.
https://en.wikipedia.org/wiki/List_of_most-visited_websites
And there is no YouTube, Facebook or Instagram (#2, #3, #4) default integration for Mozilla
Firefox has YouTube integration of course.
Here's a Firefox file [1] specifically for integrating YouTube videos into their picture-in-picture system. Your random video website won't get this treatment of course, need to be a popular one.
Here's a piece of C++ code [2] in the Firefox engine that specifically rewrites old YouTube embeds from their old HTML embed snippet to the new one. Again, your own video website will never be so deeply integrated into Firefox because it's not a top 10 website.
Firefox Readability mode makes pages more readable by removing useless stuff like videos. Unless it's a YouTube [3] or other top-N popular video website of course. YouTube videos are given special treatment because it's popular and having small integrations like this make the user experience better.
[1]: https://github.com/mozilla-firefox/firefox/blob/1f43fe5ffadd...
[2]: https://github.com/mozilla-firefox/firefox/blob/1f43fe5ffadd...
[3]: https://github.com/mozilla-firefox/firefox/blob/1f43fe5ffadd...
Twitter is a garbage fire, after all.
Welcome to 2026, where the biggest feature is the ability to turn off the AI they forced onto you.
I don't mind the AI features per se, but is there a configuration setting to sent the traffic through a local AI Gateway to prevent the AI from receiving private information? At the very least to track what is sent over the wire.
Found it: browser.ml.chat.provider
I re-tried Firefox couple of weeks ago, in the latest episode of a series of tries to "finally" migrating to it. Same CPU fans blowing like jet engines randomly. None of that with Vivaldi (which is anyway all Chromium/Blink underneath) - so came back to it.
Firefox has so many nice things like containers but basic performance issues are still unresolved.
I think that depends on what sites you are using/things you are doing with it. The only time the fans turn on for me is one specific dashboard in Home Assistant. Nothing else I connect to even moves the needle on temperature/CPU use