Post

Something that hasn't been made clear: Firefox will have an option to completely disable all AI features. We've been calling it the AI kill switch internally. I'm sure it'll ship with a less murderous name, but that's how seriously and absolutely we're taking this. …

All AI features will also be opt-in. I think there are some grey areas in what 'opt-in' means to different people (e.g. is a new toolbar button opt-in?), but the kill switch will absolutely remove all that stuff, and never show it in future. That's unambiguous. …

22 reposts164 likes

If you don't understand that opt-in for this has to mean "is not enabled by default and is never enabled except by specific user request" then your process has already failed. It has to be Off with an On button, not On with an Off button.

I've spoken to a lot of folks about what counts as opt-in. Some say a toolbar button that does nothing until pressed is opt-in. Some say the only acceptable opt-in is a build-time flag that would need manually compiled. So it's a grey area. Whereas the kill-switch is unambiguous.

Why are AI features not opt-in today? If today they're enabled by default without prompting the user, you already missed the opportunity to let the user decide. For real transparency Firefox should show a recap of all features previously enabled automatically without the user consent.

I'm not asking for faith in our direction - the thing I love about the Firefox community is how open, honest, and technical it is. But I do ask that you don't have the opposite of faith. Like, try not to be determined that we're going to do the wrong thing here. …

You already had my faith. For decades. But you've already done the wrong thing here, and that reveals Firefox has already lost it's way.

Messing with AI at all is already doing the wrong thing. It means you all care more about getting in the awful tech bandwagon than about our planet and the lives of everyone living in it. AI destroys the environment, makes people sick and kills them, steals jobs and intellectual property. Do better.

Claudie@clh2oars.bsky.social

There is *no* ethical use of genAI. It is built on stolen art, through exploitative labour, and every day as they train new model and answer query, they burn immense amount of energy & water. It has no place in our field, none.

It is also a spyware that steals our information and destroys our privacy, both things mozilla was supposed to protect. Can't stress enough how "it does not have AI" has been THE sales pitch for Firefox lately, it's what made people change from chrome to it and you all ruined it. And we WILL leave.

y'all already did the wrong thing by working w AI at all, every day millions of people scream on the internet about how sick they are of seeing it forced into their lives when they don't want it and every clown opining on its inevitability to try to keep it alive to save what's been invested in it.

there is no sufficient kill switch but to have never embarked on this path in the first place. the vast vast vast vast vast vast vast majority of people do. not. want. this. shit. the writing has been on the wall for ages and it was ignored. you want to do the right thing? walk it back entirely.

i set browser.ml.enabled to false when this started and yesterday i held leftclick on a link and it asked if i wanted to enable ai summaries. I think if you go to the effort of turning off ai in about:config it should float any other slop crap in an effort to boost discovery

if i turn the sloption off, i am never going to want to see it anywhere, and as far as i can tell that feature (the longpress link for summary) got pushed after all the initial blowback. which is why people have antifaith, and I think thats fair

We're not making assumptions you dipshits, you've straight up announced you're going to do the wrong thing. Embracing this plagiarism fueled digital snake oil is a bad decision, no matter how many options you include to turn it off

Thanks for clarifying! Unfortunately, even if I trust in the firefox kill switch right now, it is easy for future firefox to remove the switch or allow desired functions past it. The way to credibly regain trust is to never build the features in the first place, or to remove them entirely.

While deployment of the kill switch may keep me from ditching firefox immediately (the moment I saw the announcement I started looking for an anti-AI browser), it's going to take credible commitments from firefox to not do AI to keep me with firefox for the longer term.

Plus, it does mean they put in development effort, focus and time into all the ai stuff instead of other Browser features. Slows down improvements needlessly. Oh well.

As an organization, drop all plans for anything GenAI-related, promise to never touch it again, and have your wayward CEO apologize sincerely and unequivocally (and, optionally, resign). That is the only way to regain this trust.

Thanks Jake! I trust you as a person and your intentions a lot, so I hope you have enough power to steer that ship in the right directions! For me Firefox feels like a two-sided coin rn: glad Theo are here, glad they have an engine to not make Google and Apple the duopoly

But on the other side doing some things in recent years that simply don't look good :/ I guess, the issue in the end is funding. Building a fully fleshed out browser engine is expensive as hell and when your main product is a free thing for the world, it's really hard to finance that :/

Why.Add.It.In.The.First.Place? Also, why the fuck did your fucking CEO talk about it like he did? I know that you developers are not responsible for corpo shit, but the trust is ZERO right now, and your company needs to know that.

You can't with just this. Fire the CEO and roll back the AI browser bullshit, then work on actively making useful features for three or four years and ask again.

so far I can tell the dev team, that I will not return to Firefox proper. The ML was already implemented into FF without the Users knowledge and it was quite painful to disable. I did after 19 years switch over to Waterfox and don't think the new CEO can be trusted.

I don't think you or the rest of the Firefox team understand how you've blown it with your user base by giving into them "AI" FOMO and sneaking in ML settings that were auto-enabled. Take it all out!

If you want people to trust you you need to not enshittify your product. Customization to make the enshittification less severe doesn’t cut it. Build a browser that serves its users and not any profit motive, and if you can’t do it at Mozilla, do it elsewhere. Your talent is wasted there.

FWIW, I’m switching my desktop usage to Waterfox until I see some forward-looking demonstration of consistency in this approach. I don’t doubt that those working the technical side understand peoples’ frustrations but I’m tired of having to swap services because management found a new shiny toy.

The fact that Mozilla, represented always as 'for the web', is okay with even building in the most anti-web technology ever, built around scraping all our data without us getting asked (no opt in for that). Nah. Having ai even as an option is where y'all lost all the trust you built.