961
Firefox 148 introduces the promised AI kill switch for people who aren't into LLMs
(www.xda-developers.com)
this post was submitted on 25 Feb 2026
961 points (99.2% liked)
Technology
82669 readers
3335 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
@avidamoeba@lemmy.ca @technology@lemmy.world
The problem still remains: why's this thing "opt-out" and not "opt-in"? Why not make it an official, totally optional (as in voluntarily wanting to have it and, only then, proceeding to have it) plug-in or extension that the user (let us remember the meaning of "User Agent": an agent acting on behalf of the user, not a piece of software who's become "the user") could install at any moment, out of their own will?
I'm far from being an anti-AI person, I myself use those clankers on a daily basis. However, I use them because I want to, while I still want to, not because they were pushed unto me.
Mechanisms of "opt-out" where there should be an "opt-in" is a form of dark pattern.
In fact, the very concept of "opting-out" is a dark pattern per se, because it implies something pushed unto a person, something from which they were "allowed" the "right to leave".
Yeah, it's awesome to have means of "opting-out" from something, but having an "opt-out" mechanism in place doesn't mitigate the very fact that it was coercively pushed unto the person beforehand and didn't require explicit consent from the person unto which the thing was pushed.
Speaking of "consent", situations like these are not that much different from the dark pattern "Yes / Not now" we've been seen everywhere: in certain scenarious, this insistence and disregard for explicit consent would verge the criminal (e.g. harassment), but suddenly it's "okay" when corporations (and the State itself) do it.
If, say, a situation where someone is being harassed and, only after having started to harass, the harasser offers the harassed a means to leave the harassment, does this make the harasser less of a harasser? Because that's the same absurd logic behind the corporate advocacy whenever it's said "oh, but Mozilla is offering an opt-out, you can always turn off 'sponsored shortcuts' (that is, after having been faced by the shortcut from a Jeff Bezos corp as you proceeded to open a new tab for accessing the opting-out settings, but that's totally okay), 'sponsored wallpapers', and the 'Anonym tracking', and now you can, check this out, you can turn off the clankers, too! Wow, isn't that such a cute corp, the corp with the cute fiery fox mascot?".
Not to say how it's gonna end up cluttering the upstream with (more) binary blobs, adding to the Sisyphean struggle that WaterFox, IronFox, LibreWolf, Fennec, among other Firefox forks, have been experiencing upon trying to de-enshittificate the enshittificated and de-combobulate the combobulated.
"Mozilla needs to make money". Yeah, yeah, because the very fundamental, immutable principle of cosmic existence boils down to "there's no such thing as a free lunch", amirite? After all, "money" is clearly within the table of elementary particles alongside quarks and gluons, isn't it? And Mozilla needs to make money... We had a tool for that: it's called donations.
Other than link previews all the features they are opt-in in the sense you'd have to actually use the feature.
@Feyd@programming.dev @technology@lemmy.world
I'm not referring only to the feature per se, I'm also referring to any pop-up designed to appear throughout the navigation to "remind the user about the superb features".
Said pop-up is explicitly mentioned on their "confirmation dialog" upon turning off (screenshot attached below):
It speaks volumes about how much a dark pattern this is, the fact that the opt-off has a confirmation dialog, while the further proceeding with logging in with Anthropic/OpenAI/Google/Meta account doesn't seem to have a confirmation dialog.
And the fact that the confirmation feels "menacing" and defaulted to cancelling the opting-off (i.e. pressing "esc" or clicking outside the window; one must click the primary-colored "block" button which, contrasted to a grayish "Cancel" button, may psychologically induce the user into thinking "block" is a dangerous action), quite similar to the
about:configwarning screen.Ah, and the clanker options: notice the lack of alternative options for those who want a custom clanker, such as DeepSeek, Qwen, Z AI, Brazilian Maritaca IA and Amazônia IA (to mention some non-Chinese LLMs), or even something running locally through ollama. Seemingly no option for using a custom, possibly self-hosted LLM endpoint. The fact that all the options offered are all heavily corporate options (with Mistral being the "least corporate" of them all, but still Global Northern nonetheless) might tell us something...
All of these dark patterns, among others not mentioned, are the object of my critique, not just the fact that Mozilla is shoving clankers unto Firefox.
Whenever a feature needs an invasive pop-up and the opt-out brings up a second pop-up that requires further confirmation (but none seems to be offered upon actually using said feature), it is called a dark pattern, no matter if said feature requires further configuration.
I don't think it's menacing at all. It gives an informative list of features, which is nice to know. I could see a lot of people wanting to turn off all AI then realizing they actually want local translate instead of sending everything to google.
And you've got the button intents mixed up. Primary color is always the encouraged action in that kind of design. Dark pattern would be if the colors were flipped.
@Feyd@programming.dev @technology@lemmy.world
When we develop a system (I used to work as a DevOps for almost 10 years), the technical aspects aren't the only aspects being accounted for: especially when it comes to the front-end (i.e. the UI the user sees, the UX how user interaction will happen and how it may be perceived by them), psychology (especially behaviorism) is sine qua non.
Shapes and colors often carry archetypal meanings: a red element feels "dangerous", a window with a yellow triangle icon feels to be "warning" about something, a green button feels "okayish". I mean, those are the exact same principles behind traffic lights.
And signs and symbols, ruling the world, don't exist in a vacuum: a colored button besides a monochromatic button may, psychologically, lead to a feeling that the colored button is the proper way to proceed.
But... there's a twist: imagine you have a light-gray "Cancel" and a colored (regardless of the color) "Block". "Block" is a strong word. The length of the label text also does impart psychological effects. The human brain may see: "huh, I have this button which reads 'block' and it's quite strong, and this other button which reads 'cancel' and it's more easy to the eyes, maybe 'block' is dangerous". Contrast matters: the comparison between a substrate and the substances is pretty much how we're wired to navigate this world as living beings.
Now, corporations such as Apple (Safari), Google (Chromium), and very likely Mozilla (Firefox) as well, they have entire hordes of psychologists directly working for them, likely the same psychologists who'll work together with their HR departments for evaluating the candidates who applied for a job position there. These psychologists, and/or psychoanalysts, they know about Jungian archetypes, they know about fight-or-flight response and other facets of our deeply-ingrained instincts, they know about how colors are generally perceived by the human brain. Those psychologists likely played a role when a brand was chosen, or when an advertisement pitch was made. They know what they're doing.
UX/UI decisions are far from random choices from the leading team of project management engineers, it involved designers with psychologists. Again: they know what they're doing, they know it pretty well. They know how the users are likely to keep the functionality. They know how the users, as Ulrich said, are very unlikely to touch the settings, likely to keep the defaults, no matter what those defaults are. Because they know humans are driven by the "least-effort" instinct, which is quite of a fundamental principle shared among living beings as a byproduct of the "lowest energetic point" (thermodynamic equilibrium) principle.
To me, a former full-stack developer, the newer Firefox interfaces don't feel like Firefox is being psychologically fair and honest with the user's mind. Dark patterns are often subtle, and they're part of a purposeful, corporate decision.
What a load of horse shit. You don't have any clue what you're talking about and it shows.
@Orygin@sh.itjust.works @technology@lemmy.world
While some of the intricacies I brought into this discussion may stem from my specific, neurodivergent/AuDHD perception of the world, the overall thing involving dark pattern and the psychological influence of design is not something I'm inventing: it's literally an intersection between design and psychology, extensively researched by academia:
- "The Psychology of Design: Understanding User Behavior to Enhance User Experience", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 12, page no.g529-g534, December-2022, http://www.jetir.org/papers/JETIR2212681.pdf
- "The application of color theory in UI/UX design", Milot Gusia, UBT international conference, 2024, https://knowledgecenter.ubt-uni.net/cgi/viewcontent.cgi?article=4924&context=conference
- "Dark patterns and consumer vulnerability", Amit Zac et al, Cambridge University Press, 2025, https://www.cambridge.org/core/journals/behavioural-public-policy/article/dark-patterns-and-consumer-vulnerability/83EF6347CCB19EDA195C54229D34D3A8
But, you know what? Yeah, it's all horse shit, corporations don't exploit the vulnerabilities involuntarily ingrained within our brains since our births for profit, Mozilla (alongside Google who gives them money to keep Google Gemini clanker formerly Google Search as the default search engine) is a such an innocent (practically angelical) very-friendly corporation with a cute fox mascot, and I am just a pretending-to-be DevOps who clearly have no clue what I'm talking about...
The most menacing thing in that picture is the bold red text, assuming it isn't Photoshopped that way. I've seen Firefox implement other dark patterns, including hiding the ability to disable ads from within the homepage... But this isn't really one of them.
It's also true that Mozilla only supports selected AI (and search) companies, presumably the ones that give them money. Users have been begging Mozilla for StartPage integration, but Mozilla gave them a Perplexity integration instead. Firefox initially supported local LLMs in their AI sidebar, but they hid that option early on. It definitely paints all their talk about "choice" in a bad light.
@XLE@piefed.social @Feyd@programming.dev @technology@lemmy.world
I'm interacting from Sharkey, on a Lemmy thread, and you're interacting from PieFed. I'm not sure if PieFed fetches the alt-text from images. If you access my original Sharkey note, you'll see the following alt-text:
I disclosed the fact that "or pop-ups about them" was highlighted. Also, a quick reverse image search would point to the original picture where said excerpt isn't highlighted.
It would be photoshopping/photo manipulation if I removed, added or changed text from the picture, which I didn't.
Exactly, and even this one is a matter of conundrum when it's brought to the table. Because Mozilla, and corporations in general, know the exact, dosimetric approach of pushing dark patterns, not too hard so all the user base would readily notice and complain, not too soft so all the shareholders wouldn't see the "graph line go up". Just the right amount to make things dance to their song.
Even today, stating how the opting-out of "Sponsored shortcuts" isn't trivial for the average user (not to mention how said user will see the sponsored shortcuts at least once as they head to turn them off), is met with people blindly advocating for Mozilla (which, let us remember, they're a corporation with corporate interests, not a lifelong friend or a fellow trustworthy acquaintance, and corporations are driven by profit, not by friendship or psychological well-being).
The opt-out implies a feature that was pushed without consent.
Again, I bring my heavy hypothetical example: if a harasser offers the harassed a way out of the harassment after having initiated the harassment, would this make the harasser less of a harasser? Hell no, of course no! It's still harassment! It turns out opt-out features are exactly that: something that gives you the "right" to leave, only after it was pushed onto you.
And The fact that "opting-out" requires double confirmation only makes it worse, as if the hypothetical harassed were to be ask by the hypothetical harasser "are you sure you don't want this?" before being "allowed" to be freed from the hypothetical harassment.
Exactly, another dark pattern, and another proof of how Mozilla is not a friend, but a corporation.
Yeah. And this is often the justification people often use to advocate for that: "oh, but Mozilla needs to mane money" (at what cost?), as if donation-based economy weren't a thing.