Not working for me, OpenAI still exists :D
Been saying that since the start, AI kill switch should be a standart in every Application. Did that first long ago in HugstonOne, AI intelligence and speed in my own terms.
And immediately blocked.
I’m not against AI, I use it, but I want to be using it on my terms, not have it shoved into everything I use.
I kept the auto translate on. It’s the only thing I can think of that I want to just have happen.
?
Step 1. Add AI. Step 2. Add (broken) switch. Step 3. Pretend to fix switch. Step 4. Hide switch in sub-menus. Step 5. Remove switch.
… And all they actually need to do is make “AI” an extension. Let the users install it if they want to, or don’t. That’s the whole point of extensions. But they would never dream of that, hell no.
And their telemetry metrics will tell them people overwhelmingly keep the switch on.
People overwhelmingly kept autosave off. People just don’t like changing settings. I’m hoping Mozilla knows this.
EDIT: Typo
Interesting read, thanks.
Because OOBE is better than tinkering. Normies will stay on OOBE. We either tinker or move to another project.
Because the vast majority don’t read release notes these days?
I have enough trouble keeping with the IRL release notes of how my democracy is falling apart. Forger checking them on my browser.
You don’t have to read the release notes. It literally puts it front and center in your face the first time you launch it after this update is applied.
Because the people who turn off the AI are the same people who turn off telemetry
The only one I turned off was the sidebar, because that’s kinda dumb. The rest seem semi useful.
Left translate on. Everything else off. That’s one of my main use cases for AI. As long as it gets me close, whatever.
Same.
And also neither cloud-based nor LLMs
Browsers don’t need LLMs
Every single facet of my life doesn’t need LLMs. This statement is without exception.
Agreed
The only useful thing ive found for AI is its ability to read text from an image. Which is good for taking serial numbers from a photo, and copying from an app that otherwise doesnt allow copying on phone. Thats it. A tool.
OCR did that for 20 years .
Nothing these slop generators do is novel or new.
I remember using Google translate that was doing that live on the phone camera and translating the text at the same time 15 years ago.
Random aside to rant about consumer OCR.
Recently for my work I had to do some OCR stuff to get some numbers out of a document that the vendor in their infinite wisdom refused to provide in an editable/selectable form. I.e. they just slapped a .jpeg onto a page and saved it as a .pdf. (This is a separate thing that infuriates me.)
Anyway, what I’m actually here to complain about is the baffling phenomenon that every single piece of OCR software I tried ranging from open source to trials of commercial programs, to the thingy that came with one of our all-in-one printer/scanners, and everything in between is that it’s somehow still exactly as crap as the lousy OCR programs we were all struggling with in the late '90s.
I have absolutely no idea how this facet of technology in particular has utterly and categorically failed to make any forward progress whatsoever in literal decades. I’ve personally worked on machine vision driven pick-and-place machines capable of accurately determining the orientation of densely printed cosmetics tubes, among other items, and placing them all face up in a box several times per second. Yet somehow the latest and greatest OCR transcription algorithms still can’t tell a 5 from a 6 or ye gods forbid an S, or an L from a J, or an M from a collection of back and forward slashes, all despite being handed crisp high contrast seriffed text that’s at least 60 pixels high.
Given the incredibly low bar for performance here given that apparently every single programmer involved just walked away circa about 2001, I can’t imagine that the current slop generation machines fare any better…
I have tried some of the popular LLMs a few months back when I had to digitise an old policy document from which only an old scan still existed. I had trouble reading it.
The results varied wildly. OpenAI was really poor at it while Gemini got it right completely. I was quite impressed. ABBYY FineReader is supposed to be the best non-LLM software for OCR, but it doesn’t come near the performance of Gemini
How else do people think we were translating all that hentai before the slop generators took off
OCR kinda sucked lmao
Always worked well enough for me.
I remember trying to use some pre-LLM OCRs and it often got hand-writing really poorly. LLM backed seems to perform generally better, now typed OCR was usually pretty good.
Yeah, I never really used it for handwriting. That seems basically unverifiable sometimes, when I can’t read it myself.
that function is just reskinned OCR, though
which I guess you could consider as AI and that it is a similar training data structure? not my area lol
I do also think that AI has some use as a search engine. I haven’t used it much for this purpose at all, but a while back there was a specific type of engineering analysis I needed to do, and I couldn’t remember the exact terms or topics to look up. chat GPT got me into the right area so I could look at the appropriate resources. in that specific scenario, it was better than a standard search engine
Of course once I found the materials I was looking for, I stopped using the chat bot and you know use those materials
Yeah, ocr is a type of AI. The big advantage of modern techniques is that it can factor in context a bit better. It’s the same principle but a different mechanism for how you know a red hexagon with S__P on it says stop, even if the sign is dented, a letter fully fell off, it’s raining and dark.
It also means it’s sometimes wildly inaccurate, like in cases where it’s just so much more likely that it said something else. Like how on a bright sunny day, with perfect clarity, and a crisp new sign with extra good visuals, you’ll hit the breaks for a sign that’s a red hexagon that says §¥¢¶. It’s just very unlikely that that would coincidentally be on a red hexagon near the road, so it’s more likely you saw wrong and it was actually the normal thing.
Ackshually… Stop signs are octagons!
I also find LLMs decent for translating text between languages, though for serious use it still requires human review
“Kill switch” – oh the drama. Let’s call every simple toggle ‘kill switch’ from now on.
THERE’S A PHYSICAL KILL SWITCH FOR TYPING IN ALL CAPS ON MY KEYBOARD, BUT I HAVEN’T ACTIVATED IT YET.
“Firefox is including an AI murder switch so that heartless users can take the life of our helpful little robot guy who just wants to see you happy. We added it because not everybody is a good person.” -Mozilla CEO.
i have a violently execute switch in my room (it toggles the lamp on or off)
For it to be a kill switch it would have to actually terminate a rogue AI.
Yeah, call me when Firefox creates terminators that infiltrate and destroy data centers and then themselves.
What’s worse…you could always toggle it. In fact, you could re-route it to your own local LLM.
Drama drama cheesecake drama
Tablet brick potato drama
I was once a treehouse; I lived in a cake. But I never saw the way. The orange slayed the rake.
Are… are you the fabled walrus? Goo goo g’joob?
Or maybe you should just get over your superior self and let people have fun.
The only people that are into LLMs are scientist (which is reasonable) and tech bros.
The later just think it’s useful while for 99% of people there just isn’t a usecase.
I guess I’m in both groups. I love LLMs, but then again I have academic degree in AI.
Though I must also admit - look how they massacred my boy. LLMs could be used in games to make every NPC a talkative character or work as a customer support during off-hours of small businesses… instead they are used to generate advertisements faster.
Having tried a game with AI NPCs, they’re fucking awful. I don’t want to sit there and have a conversation with irrelevant individuals. Give me that concise “here’s your mission and reward.”
An AI companion who can converse, comment and learn may be cool, but I definitely don’t want anything like what I’ve seen.
So while previously the translation feature was supported by an extension, now it has to be enabled through ai.
Hate it.
Im super happy to see so many upvotes for this most excellent browser!
Remember when they had a “kill switch” for javascript?
Yeah, as long as it made any sense to browse the web without JS. These days you need at least an allow list.
It’s available as an extension: https://webextension.org/listing/javascript-toggler.html
Yeah. Used to be native. Like the slop kill switch currently is. Then won’t be.
For years and years I’ve been using this extension: https://addons.mozilla.org/en-US/firefox/addon/yesscript2/
Mozilla has released so many self-described AI features in the past few years, but this is the only one that has:
- been requested by the community
- received broad critical acclaim
I hope Mozilla learns their lesson. I doubt they will, but I hope.
To be fair people liked the translation feature too
I tried FF translation a couple times, and it’s woefully poor compared to Google’s. What am I doing wrong?
Firefox’s runs locally while google’s runs on their (much more powerful) servers, for something similar to chrome’s I’d just get the deepl extension, which does the same thing just better.
TWP does it better.
TWP dumps your pages to google translate, no thanks. FF is on device
You don’t have to use google translate (there are 2 other services included), and TWP doesn’t reload the page when you toggle the translate function off and on like the built in one did.
Those are still external services. I didn’t mention them because they carry the exact same risks
Ssshhh don’t say that too loud or the “no one wanted this” crowd may hear you. They would be very scared if they could read.
In the original announcement that they added translation, they didn’t call it AI. They didn’t even call it machine “learning” or machine translation there.
They just called it local, automated translation.
Maybe you should take your own advice about reading, and double-check my comment ;)
When I turned it off the translation thingy went away, so I’m not sure if it was AI all along and they were lying about it or not. Just as well, there’s an extension that works fine and it doesn’t reload the page every time I toggled it like the built in one did.
The translation is technically AI, but it’s a distant cousin to the LLMs and image generators that have repulsed so many people. (The term AI is such a broad and vague umbrella that Netflix recommendations count as AI.) And, even more notably, this is before Mozilla started marketing things as AI.
It was also a joint non-profit venture with a university, rather than today’s weird gimmicks or for-profit partnerships.
It’s less a vague umbrella and more an academic category. It just feels odd to call it vague in the same way you wouldn’t call “chemistry” vague, despite it having applications ranging from hand soap to toxic waste.
In this case, the vagueness of the term AI is abused by its fans. “Aha, you claim to hate AI, and yet…” they say. They should know better.
“Chemicals” is actually a great example. If someone said “Chemicals are coming out of that factory”, you’d rightfully cringe if a factory manager said “well actually soap is made of chemicals too”
And?
Because the term AI was not in vogue at the time, even though it’s clearly the same technology, it doesn’t count? It’s literally packaged under the same umbrella now.Anyway, the big issue is still tech ppl thinking their viewpoint is the only one valid, and that every generic user will have the same exact needs as them.
I already addressed all of these arguments in another comment in this thread…
Not all these arguments no.
You’re defending your position that this AI feature is not really AI so it’s ok, but the others are all bad because of the two letters of the devil.
Still AI is a marketing term, always has been. AI in the form of machine learning has been around for more than a decade, and lots of things already use that.
The knee jerk reaction of tech circles saying mozilla will sell their soul because there is no “kill switch” is so fucking dumb. Even more dumb is thinking no other users may want any of these features. Unless you work at Mozilla, and/or do product research for browsers, chances are you most likely have no idea how people will want to use these features in their day to day.
Even working on one’s own product in a company, few really understand the users needs and wants, especially tech persons.
I can guarantee you, the weird gimmick you don’t understand is crucial to some.You’re defending your position that this AI feature is not really AI so it’s ok
I literally say “The translation is technically AI,” so no. I give reasons how the other features are different, which you seem to acknowledge a little, at least.
the weird gimmick you don’t understand is crucial to some
Can you describe how to access the gimmick and which people find it crucial? I’m pretty confident in my understanding of it and how hilariously unhelpful it is.
sadly I’ll likely support them through any shitty decisions they make as they are the only viable non-chromium alternative these days.
I get they’re chasing the buck and trying to stay relevant, but uhhhh… if they could be less Steve Buscemi-teen about it, that’d be great.
I strongly believe that the EU should fund Mozilla, or a fork of Firefox.
Gecko is the only viable competitor to Blink/WebKit, and it is needed
Govts around the world should be funding all sorts of FOSS projects. I know they do to some degree but not much. It benefits the whole world and only hurts big tech.
That prospect becomes less and less likely the more government is bought and paid for by Big Tech.
This is what people don’t understand. Those in power, whether they’re part of the government, a wealthy CEO, or a religious leader, will do what benefits themselves if they think they can get away with it. We keep talking about powerful organizations and what they could do to benefit everyone, but fail to realize that powerful people don’t want to benefit everyone.
They only do what benefits everyone if they feel like they can’t get away with just doing what benefits themselves. It’s our responsibility to make sure they don’t think they can get away with it, and clearly strongly-worded letters and quippy signs held outside their offices for an afternoon or two isn’t enough to do that.
Funding FF? Maybe. Funding Mozilla? No way, not with my money.
Why?
Firefox is just the browser, Mozilla is the organization constantly wasting money on features Firefox’s users are actively hostile to in a bid to tempt away people already using Chrome. Not the OP, but I’d be down to donate to Firefox’s development directly, but I wouldn’t want to make a donation to Mozilla hoping it would go toward Firefox, only to find out they took my money to build some new LLM integration that nobody asked for, only to sit unused for years before being quietly shuttered in favor of the new tech buzzword of the day.
Yeah I really hope there will be some way to tie donations directly to FF development.
Maybe funding components would be better than funding mozilla. Eg: 2 engineers for Gecko
This is probably common knowledge to you and many others, but it bears repeating: You cannot donate to fund the development of Mozilla Firefox.
Google can, unfortunately.
I’m pretty sure you can donate to Mozilla, are you saying that donations to Mozilla don’t accomplish what we might want? I only know a little about how much Mozilla sucks so I’m ready to learn more.
There are two interlinked Mozillas: the Mozilla Foundation is what you can donate to, but the Mozilla Corporation is what develops Firefox. No matter how much you donate, that money cannot (or will not) be transferred to Firefox development.
Mozilla’s recent history includes privacy flub after financial management misstep, so here’s a few links in no particular order of severity or chronology:
- https://krebsonsecurity.com/2025/02/nearly-a-year-later-mozilla-is-still-promoting-onerep/
- https://techcrunch.com/2024/11/05/mozilla-foundation-lays-off-30-staff-drops-advocacy-division/
- https://smoshed.substack.com/p/ublock-origin-lite-maker-ends-firefox
- https://www.coindesk.com/business/2023/02/07/crypto-wallet-security-layer-webacy-raises-4m
- https://mastodon.social/@stevetex/113162099798398758
Each of these are a little rabbit hole on their own.
Reminder that Krebs is an asshole who responds to criticism by doxxing the accounts of other security researchers.
I recommend Waterfox
They have pledged to not fill their browser with AI slop features.
Last time I tried Waterfox some sites like Twitch that actively block usage on old browsers, refused to work because the latest Waterfox release was based on a Firefox like 20+ builds behind.
Firefox was on like version 142 and the latest Waterfox download was based on build 128.
Waterfox right now is built on ESR 148, which is on par with the latest Firefox release! ESR releases will lag several versions behind, but that’s normal (even on Mozilla’s side), and I’d be kind of shocked if it was such a big gap
Edit: there was a big gap. 128 to 140 was the right jump, but Waterfox non-betas took a little less than two months to implement the change after Mozilla released it.
I have used it for twitch for years without issue. I also have ublock origin with twitch adblock.
Well you clearly haven’t used the standard available download (non-beta/nightly release) consistently through last year. Waterfox was using ESR 128 since October 2024, kept that base until finally upgrading to ESR 140 last August. So that’s nearly a year of its base being out of date. So the user agent reported that number… sites really don’t like that since they’re looking at that for support.
https://www.waterfox.com/releases/6.5.0/ https://www.waterfox.com/releases/6.6.0/
Twitch only supports the last TWO versions of Firefox officially and will actively block logging in from older versions. So while you might be able to watch Twitch, if you aren’t already logged in, you won’t be able to login.
https://help.twitch.tv/s/article/supported-browsers?language=en_US
There are thousands of posts about it online for Waterfox other forks.
It was outdated, but only for a couple months. Firefox ESR is built to last about a year, and it was maintained with security patches up-to-date alongside Firefox Production versions 129, 130, 131, 132… all the way to 139. Only then did ESR 140 come out.
But if Twitch only supports the two most recent Firefox production versions, I guess ESR wouldn’t cut it after FF 131 came out.
I have used the standard available download on multiple operating systems for years without issues with twitch.
Quintessential “works for me” response. Must be a software developer.
If everyone switched from firefox to waterfox, Mozilla would kill firefox which would in turn will waterfox
Yeah ofc they are chasing the buck.
It’s either they find alternatives revenue streams or we no longer have Firefox as a viable alternative anymore.
Browsers development is crazy engineering heavy, and thus, expensive.
It’s a shitty situation all around.
Ladybird browser looks promising!
The ladybird devs are currently in the process of switching language again from Swift to Rust, using LLMs.
Yup. Don’t use or support Ladybird, especially since it’s made by anti-inclusivity “keep your ‘political’ gender-neutral pronouns out of our READMEs” nerdbros.
On the other hand, Servo is coming along nicely.
https://github.com/LadybirdBrowser/ladybird/blob/master/CONTRIBUTING.md
Use gender-neutral pronouns, except when referring to a specific person.
?
That line is actually part of a very interesting change to the document.
https://github.com/LadybirdBrowser/ladybird/commit/627dcb90bdd23ccfa2ae210d55b474ab4a844db0
Problem is Mozilla needs money and shoving AI features into shit is how you get investors these past few years.
You think VC is putting money into firefox? Wtf?
Funnily enough, it’s the other way around: Mozilla has been dumping money into AI VC startups.
I don’t think the vietcong are doing much of anything for the past couple decades.
I think they’re desperate to make money since they’re losing userbass AND Google is probably not happy that most users change the default search engine away from them.
Does anyone really think the current administration is going to break up Google? Lina Khan almost did it but like most of the rest of this timeline we just didn’t quite get there
Yeah it’s a catch 22.
They either fail to get a big enough use base because their core users are not enough and they fail from a lack of funding.
Or they try to follow trends to increase their appeal and user base, and annoy their core users.
Most users don’t realize that Mozilla is doing what Google is doing with Chrome with an engineering team 1/4 the size of the chrome team. And that the grand majority of their costs are engineering related.
Browsers are expensive, and Mozilla needs to find revenue streams to pay for it.
I believe Firefox could raise a lot of money through donations. If they make it clear that Firefox donations will be solely used for Firefox development. Also ideally add a quick survey to donations to see what the “donating” userbases values are. My issue with donating to Mozilla is that it is too broad and they have many products I don’t care for.
I use Thunderbird and donate to it because I feel it’s more focused. I believe Mozilla still can use the funds for other stuff but at least I am donating for a clear project.
Firefox donations will be solely used for Firefox development
This might be a stupid question… but how much developing does a browser actually need? I get security updates and such but how much resources does that stuff really need? Full disclosure: I’m a dumb lorry driver I have no idea how these things work. Some years ago I realized I hadn’t updated my browser in at least a year, maybe two and I had no issues lol
A conservative guess would be around 60 people.
https://bugzilla.mozilla.org/describecomponents.cgi
You can click around and see the bug reports they’re working on. There are a few, to say the least.
https://www.firefox.com/en-US/releases/
This is a way to see what’s in each release. The ones on the left are major releases and tend to have bigger features, and the others tend to be bug fixes.
Web browsers start with core functionality that’s very complex. Then you tack on that they’re being used for things like banking, and managing the critical details of people’s lives. That means security galore, which is hard and constant. Then you have ad people, who are also something that’s hard to defend against.
Then there’s the constant flood of new features you have to implement to keep up with Google.Chrome has 1,000 to 4,000 people working on it. Mozzila employs about 700 to work on firefox, with maybe 1,000 additional open source developers.
My initial guess was very wrong.
It is really difficult to implement in the first place, and the standards evolve constantly.
Some argue it may not be possible to build new browsers anymoreInfinite money Google keeps trying to push shit to the standard so all other browsers end up needing significant dedicated resources to keep up or risk getting blamed for broken sites.
I, the laziest man possible, have been motivated to switch already. Waterfox is working just fine.
So, there’s a “bug”, though I expect to FF it’s a feature: If you individually block all of the AI features, then click on the master switch to block all AI, everything’s great. But if you revert that master switch suddenly it “forgets” all of your settings and shit is activated again.
It seems by design. And since it’s opt in, if FF “accidentally” disables the master switch (I’m betting it will eventually) you lose that extra layer of protection. OH, and I had disabled EVERYTHING in registry (about:config) before this and translations were still available. I guess it’s time for me to explore other FF-core options…
Lmao, semi common design mistake? MUST BE INTENTIONAL!
I don’t think I’m being paranoid by saying it:
-
opt-out rollout of every AI feature
-
only slogging through registry to manual opt out until now
-
CEO and board hell bent on monetizing and delivering features users actively do not want. I.e., enshitification
-
I have seen my own AI registry changes revert already once after a patch
-
It’s just a lazy/poor design.
Instead of each setting having its own bit with one ‘override’ bit, they just set override by setting each bit.
I’d say you’re being generous calling it poor design. It’s actually reverting to “default” on settings when you uncheck instead of storing individual bits and honoring those. Why not revert to opted out - OK, that may be lazy to use a single template, but that’s not the way some of their other “master” options work. And I’ve been a FF user since it’s first releases, so this isn’t some Mozilla hate. And I won’t be going to anything Chromium and because of inertia I may just stick to FF.
It’s also crazy that I have been manually configuring away from AI since it wasn’t even opt out… it was forced in. Most aren’t going to do that and Mozilla knew it going in. And I’ve already seen those registry settings revert once. Since this control option literally should have been the first feature for AI delivered and their entire AI push has an untrustworthy stink, I’ll say it again: I await a future release bumping the setting back “on”. “Oopsie! you can just turn it back off or wait for the next patch” after Mozilla and their partners collect their information across millions of users that aren’t paying attention.
Wow some people really hate AI huh. Did a robot kill your father or something?
