It seems to me that adding AI to desktop apps and sending the data back to the mothership for processing is an amazing way to collect data from people who, for the most part, would be completely unaware it's even happening.
> Heck, most of them think the Internet is Chrome.
In the end Google has achieved something that Microsoft couldn't with Internet Explorer, and won the Browser Wars.
Google managed to aggressively advertise their browser by optional install "offer" within Windows installers of software. And they were aiming exactly at all those who couldn't tell the difference between the web browsers and who were conditioned by more experienced family members, friends etc. to just blindly click "Next Next, Finish". Thus, that was an easy win.
Being here when we had choice between Gecko, Presto, Trident and later WebKit/Blink makes me sad how easily the IT world allowed this nearly 100% monoculture to happen. There are still other browsers but chances that we return to variety and choice of rendering engines are low.
What annoys me the most about this is that we as web developers have been here before with IE and managed to get out of it. Then we immediately forgot that lesson and helped Chrome be the new IE in a new coat by adopting ”modern web API’s” that were in fact proprietary Chrome API’s with low respect for the standards process, until every other browser was forced to accept it as a standard. All those ”Safari/Firefox are lagging behind” or ”best supported in Chrome” wasn’t in our best interest or even, from a standards perspective, particularly true.
Every vendor tripped over the third party affiliate human review issue.
While consumers remain surprised by affiliate clauses, the QC problem is considerably different from marketing against those recordings.
The linked article veers into Alexa for the ads part and says, roughly, must not be tin foil hats if everyone believes it – then explains the psychology misleading people in most cases. The "I'm noticing a pattern" thing…
Are there sources where Apple either acknowledges or even settles claims of advertising against secret Siri recordings?
> Are there sources where Apple either acknowledges or even settles claims of advertising against secret Siri recordings?
Yes, advertising and selling data to third parties.
> In January 2025, Apple settled a class action lawsuit brought on behalf of current or former owners or purchasers of a Siri-enabled device, specifically: iPhone, iPad, Apple Watch, MacBook, iMac, HomePod, iPod touch or Apple TV, whose confidential or private communications were allegedly obtained by Apple and/or shared with third parties as a result of an unintended Siri activation. (https://www.nbcnewyork.com/news/apple-settlement-95-million-...)
They've allowed third party contractors to listen to apple customers fucking (and who know what they did with those recordings). Apple's privacy policy says that they'll share (which includes selling) your data with "Apple-affiliated companies" and "our partners" without disclosing any of them.
I've seen claims that they've sold user data to facebook, but I haven't seen any details except the existence of "user-data sharing agreements" between apple and facebook and of course they take millions in cash to hand apple user's data over to google through their search engine.
Apple has been careless about handing data over in response to legal requests (https://www.cnet.com/tech/services-and-software/apple-facebo...) and gives a ton of user data to governments in the US and china. According to Snowden Apple partnered with the NSA as a data provider. Apple denied that. They've also denied working with the US government on DROPOUTJEEP.
They honestly don't have a very good track record when it comes to privacy, they're just quieter and less overtly evil about it than Google, MS, and facebook.
Something no one else would want - a little colored dot next to HN user names keyed to their generation, so I could quickly tell why none of the commenter mentioned AOL.
Great joke, but taking the word literal and not as product name, it makes a lot of sense to describe chrome as a tool to explore the internet.
(Edit: thinking about it, I think generic terms like "Internet Explorer" should not be trademarkable at all, also I just learned, that also Microsoft "stole" the name and had to pay in a settlement..)
No, I think it was meant literally. Like the IT Crowd skit where Jen refers to the Internet Explorer desktop shortcut as "the button for the internet".
According to the current shortage of computing power and electricity, I suspect that what they really want is not your data, but the computing power and electricity from your device.
If users' behaviors can be pre-labeled on their own devices, processed with AI, and then sent back, it might save a significant amount of internal computing costs.
Chrome does/did this already. They've never been of any forensic interest to me but the history file stores clusters of domains visited and keywords searched for. It was pitched as user "Journeys." I think they've deprecated it.
I called this out when it was announced on here. Supposedly the team lead replied to my comment saying this wouldn't happen. I rolled my eyes but asked will android be able to use those models for ex filtration. No reply. And apparently the original claim was not true either lol.
Maybe I'm misremembering it. Google is awful. My goodness. I hate Android and can't wait to be rid of it. Graphene and it's buddies can't roll it fast enough
What you should've done is saved what they said so you could post it directly as evidence. If they're collecting all the data they can, you should naturally also have the right to do the same! I've noticed they're increasingly memory-holing a lot of things that, somewhat coincidentally, are inconvenient truths.
It's in my comment history..sometime last week. I don't really care, I knew this was a surveillance move that's why I called it out. No clue if the anonymous poster was who they claimed to be anyways
Does anyone believe a single big tech company isn’t harvesting data en masse from everyone in duplicitous manners?
Like, the best case scenario is that they don’t just blatantly steal your data and instead use dark patterns or inference to take from you without your knowledge.
And then, thanks to the wonderful opinions of the court, the government has full access to said data since you apparently knowingly agreed to giving it to a third party by virtue of the fact that you engaged in any sort of commerce.
It’s why I’m for forcing content being posted on the internet to be non anonymous and tied to a real identity.
The corporations and government already have and abuse all this data. I want the benefit of knowing when someone says “As an American {incredibly divisive shit}” that it’s actually someone in a foreign country sowing chaos for money or political aims.
They won't actually show you who said what though. Twitter trialed that feature then they quickly turned it off after everyone realized half of the maga influencers were russians.
It also kind of stinks because not every mistake should be immortalized and recorded forever. Blackmail and all that. It kind of ruins the internet in a different way.
I want it enforced at a governmental level. I don’t think content consumed by people should be required to be public but if you want to post on a public square that the internet is, I should be able to recognize you as well I could in meatspace.
We get all the negatives of anonymity now with foreign actors, domestic actors, and bots flooding the zone with lies, and not of the benefits since all the corporations and governments can trivially pierce that veil.
I assure you they cannot pierce that veil if one has a modicum of competence.
One uni in my country has been getting bomb threats during the exam period every year for multiple years (a random article says 20 times at least). The whole place gets evacuated each time, nothing is found and nobody is caught.
But people who think they're anonymous because they used a different nick? Yeah, those are idiots, their ISP and the platform knows who they are and anybody can deanonymize them through stylometry.
I don't think surveillance is the solution though. I'd much rather see a network of trust or (second best) anonymous proof of identity.
Any place selling alcohol or cigarettes is able to check if you're 18. They could just as easily check your nationality by looking at your ID and give you a crypto key which can be used to prove that to online platforms without revealing who you are.
But there's no money for big corps in that and most people are not even smart enough to think of it.
Every single packet of information sent on the internet flows through NSA servers at some point. We've known this since Snowden. When I say the government has this information I do not mean that every single bureaucrat at any level of government can access this, but that the government as a whole has access to the information if they feel its worth using. We have limitations on how the government uses such powers but those are currently worth the paper they are written on and being ripped up left and right.
That's defeatist, although certainly what they'd like. The world is not the US.
Even if NSA captured and logged all internet traffic, they'd still only get a fraction of the information within without breaking all the encryption.
And even if they could break the crypto, the ability would only have any power if/when they acted on it. Which in turns reveals the capability to both normal people and other nation states.
The limitations aren't laws, it's the practical consequences.
No, when they can get away with it. When the pros for them outweigh the cons. And every person affects that calculation in some small way - by their tolerance to injustice, by their outspokenness, by their ability to detect it, by the probability of an employee becoming a whistleblower and even by people's willingness to punish bad actors extra-legally.
> They still have the ability to do it at will.
Yes and your neighbor has the ability to wait on your doorstep until you open and then slit your throat.
He generally doesn't because of internal safeguards (humans have evolved emotions like remorse, though not all of them feel it - see psychopathy/ASPD for details) and external safeguards (the state generally opposes people infringing its monopoly on violence but in the absence of a state, he would still risk being hunted down and tortured/killed as punishment by your relatives and friends).
---
I get what you're saying and I oppose dragnet surveillance but seeing things in pure black and white does not help.
Beyond the easy quip, my point was that "big tech" has taken on a far-too-narrow meaning. I'd bet you didn't really mean "software" either - you were, in effect, referring to the whole amoral (at best) ecosystem of companies which run social media and related web sites and data infrastructure, making their money through addiction, exploitation, and extraction.
Perhaps Cory Doctorow will come up with a better term?
No, I meant companies that have a lot of money and are primarily software. Discord for example is a company that isnt as big as a faang but one of the companies I would have included in "big tech",
The software companies, for whatever reason, have been largely bad actors when it comes to dealing with users data
I'm just not sure how this gives me control of my information, whether I want it sent or not to Google, and if they're retaining it for training or not.
That last question I don't even want to ask because the first two doesn't seem clear.
This could be simply fixed by adding the feature, and defaulting it off, and letting people learn about it and enable it.
> I'm just not sure how this gives me control of my information, whether I want it sent or not to Google, and if they're retaining it for training or not.
That's not the goal. Turning your information into their information is the goal. If training an LLM on some data isn't copyright infringement but instead makes it a brand new non-derivative work, then training an LLM on your personal information arguably means it's no longer your personal information, but instead a legally distinct work.
I'd argue that even pre-AI the average internet surfer never thinks about all the data the sites they use collect. I'm not even sure if my mom uses any apps that aren't web apps (maybe MS word).
But for it all to go to one place? That's a scary amount of data.
The even more frustrating thing here is that after auto-updates everything new [including AI "features"] is turned ON by default.
I do like how Firefox now has a "prevent future AI integrations" checkbox[0], but I just don't believe it anymore (i.e. that it won't magically `uncheck` itself and then enable features I've not requested/authorized).
Which is why I just used an LLM to help me create a local network admin rule to disable the update engine entirely (this SHOULD. NOT. BE. NECESSARY).
My belief is that the AI business is all about data collection. The value isn't so much in the quality of the models (that's what enterprise customers and developers pay to get), but in the amount of data that comes "for free" to whoever hosts the models. And then it's worth whoever buys it thinks it is, like insurers or advertisers.
It’s because they’re all in bed with the Government who wants the same thing. Think of them as an extension of the NSA/CIA and the byproduct of the data collection is being able to sell Ads to make money.
My guess: people would ask way more for that kind of information. That's why it's a totally immoral business: making people give away something they would never agree to give for free, if they were truly aware of it.
it's more like knowledge extraction at this point. younger generations don't build up knowledge any more, everyone else is slowly losing their knowledge by not using it.
Eventually the rug pull comes and knowledge will only be accessible by those who can afford it.
Up until ~2 weeks ago, I believed that at least opting out of data collection would protect me. I no longer do.
Everyone knows that fines paid by companies (instead of the people making the decisions) are considered simply a cost of doing business. A probabilistic tax, if you will.
What finally dawned on me is that given they need more and more data to train bigger and bigger models, at some point the value of using my data for training will exceed the cost of getting caught using it without/against my consent.
Yeah I was wondering how long it would take for a browser company to do something like this. It lets them scrape data without having to deal with anti-scraping provisions on websites, since now their training data collection gets spread across the entire Chrome userbase and they're able to offload the work of bypassing the Cloudflare captchas or whatever to their end users.
When you use a coding model running on someone else's computer, you're giving an AI company your proprietary source code and associated documentation, and you're giving free training examples to make a future AI model better equipped to eliminate your job. Valuable data indeed.
Which is why it is odd to see so many companies jumping on the bandwagon, even those that have always been super protective of their oh so valuable proprietary internal code.
> My belief is that the AI business is all about data collection.
In the short term, maybe. That's what you tell investors.
In the long term, it's about altering, shaping, and even constructing reality: making a new and canonical truth for humanity where the ruling classes are invisible to us and the machine that tells us our history and bedtime stories and how we feel is in every device we carry, until it is everywhere, and it has always been everywhere, and it will always be everywhere.
Yes. It is seriously not a coincidence that all of the ai companies are now offense contractors for the department of war. It's also not a coincidence they want to ban vpns, and force people to verify themselves with IDs, biometrics and their phones for all of their activities. Meanwhile... Bots can run free.
This seems somewhat specious - it's also quite possible that they just altered the wording to make it less verbose. Does anyone have access to the link "Learn more about on-device AI"?
If Chrome starts sending data from the browser back to Google, that's going to be a huge compliance issue. If you work for a company that processes customer data in the browser, you're going to need to ban Chrome.
Chrome has been recording metadata (URLs, timestamps, etc) about your activity since forever, and you can turn this off if you like, see https://myactivity.google.com/myactivity.
> Some AI features in Chrome do not rely on on-device Generative models, and those features may still run even if the on-device Generative AI models are removed.
it already sends data back to google, the ai stuff, everything that goes in the address bar goes straight to google unless you specifically configure chrome to block it
the on-device ai just offloads some work onto your device
i doubt anyone will be banning chrome, for some reason "it's for ai" is a valid excuse for any amount of sillyness
Yeah, I staunchly refuse to believe an ad company that releases a closed-source browser would violate our privacy. You're probably right that they changed the claim simply because it was too verbose. That's the best and only explanation.
I know that I'm in a bit of a bubble with this one, but I am surprised there is still anyone using Chrome instead of Brave. I get the dependency on Gmail other Google-specific tools, but the built-in ad blocking and Google-free aspects of it made me switch instantly and haven't look back after years.
Chrome has stayed incredibly sketchy from the beginning, when Google gained marketshare by sneaking Chrome into the installer for other products that people intentionally downloaded.
Then Chrome did things like "accidentally" uploading your entire browsing history to Google servers when you signed into Gmail.
Now they have declared war on ad blockers, despite the government warning that ad networks are too big a malware vector to ignore.
All they did was add their own affiliate link to crypto links that didn't have them. You didn't get tracked from it, and you didn't lose out on anything.
Still sketchy because of the lack of consent, but people act like Brave personally stole money from them.
The other "sketchy crypto stuff" is one of the few actually workable alternatives to funding websites with ads on webpages. Again, Brave took in no money (BAT) that you as an admin / creator would have otherwise had, and they keep it in escrow, they don't claim it.
The only other sketchy thing I can remember is pre-installing a deactivated VPN so that people could pay, push a button and it'd work immediately. Plenty of companies do hacks like that for the sake of UX. Dropbox used to hack macOS its Accessibility permission so people wouldn't have to dive into settings to toggle certain things.
The irony is that Firefox has had their own scandals like surreptitiously installing a hidden ad extension that would advertise for Mr. Robot, but somehow Firefox stans have erased that kerfuffle from their collective memory.
> The irony is that Firefox has had their own scandals like surreptitiously installing a hidden ad extension that would advertise for Mr. Robot, but somehow Firefox stans have erased that kerfuffle from their collective memory.
Brave is my default browser for non-sensitive tasks; e.g. most web browsing, GitHub, news, etc.
The built-in ad-blocker & tracker blocker alone is worth it.
Use chrome for testing.
Stock Firefox for anything sensitive.
In my mind, no browser is perfect. However, as far as I can tell that’s not nearly as sketchy as the title implies. It’s for local debugging.
Zen has other issues for me on Ubuntu (eating a ton of resources) which is why I usually use FF. But I put Zen in a different category from Brave and definitely better than Chrome.
I switched to Firefox when Chrome started messing with the ad blockers. Haven't really had any issues. I prefer developer tools on Chrome but I rarely need to use them anyway.
The trouble is that Mozilla has admitted they can't survive without Google's revenue. You are basically using Google by proxy unless you use a truly independent browser engine of they get blocked by Cloudflare for not having enough fingerprinting tech.
(Ungoogled) Chromium and Firefox are both projects that are open source and readily available. The code is sitting there ready for you to compile. More users = more donations. You can be the change you wanna see.
> You are basically using Google by proxy unless you use a truly independent browser engine
This conclusion doesn't follow your premise. Google has to pay because if Mozilla dies, so does the claim of any real competition on the browser engine market. So everyone agrees Firefox's engine is truly independent. Google pays so Firefox users don't use anything that has to do with Google.
If you think about it, the only real way to not hurt Google is for Firefox to stop existing. Chrome would end up being spun off from Google.
I don't agree that you are using Google by proxy when Firefox has more technical independence from Google than Chrome and can be quickly decoupled from the few Google defaults it has, search and safe browsing.
What browser can genuinely claim independence from Google? Chromium browsers are all arguably in the same camp. If FF is implicated, then so are forks like Zen.
Because some things only work in Chrome. It's a fact. It's terrible.
We're the frogs being boiled, over the last decade. People sounded the alarms, but they were looked at like they had tin foil on their heads. Now, it's clear they were right.
I'm speaking generally, of course. I use Firefox for all my personal stuff, except for those situations where it doesn't work.
Chrome likes to make up new "standards" and then some websites adopt them immediately.
That said, I can only remember two instances of that slightly inconveniencing me in the past, and both times I was inconvenienced by a Google-run website: once upon a time Google Earth refused to work, and once upon a time I couldn't tweak my Google Meet background. Both are no longer the case.
Citation needed. I've seen the opposite--unless there's a very specific niche that can't be otherwise solved, there's huge internal resistance to going it alone.
The biggest counterexample I can think of: WebUSB was critical to Chromebooks supporting external devices, but I can see why Safari might not want it. It has Firefox support at last, though.
Citation of what exactly? That not all browsers implement the same thing at the same time and that some features are Chrome-exclusive because for one reason or another other browsers refuse to implement it?
Is that really something you need a citation on? You sure seem to have come up with an example of your own.
In the past there were features that didn't work at all; I used to hit those regularly. Device setup flows, AV features, etc. These days, it's never "this doesn't work on other browsers". It's always "this is worse on other browsers", whether because they don't test it or because they don't care.
YouTube is terrible on Firefox. There was a period where it was usable but got increasingly worse with missed frames, low frame rate. On FastMail and Gmail the expanded search overlay doesn't disappear when you click outside (ESC doesn't work), you often get stuck with it. On YouTube when you stop hovering over the "I like this" etc. on full screen video view, the tooltip doesn't disappear. It's death by a thousand cuts.
I wonder if he's using a Radeon gpu, those have had issues like video playback making the entire ui framerate drop. But it happens in steam/electron too, not only firefox. The shader cores aren't used much either, it's just a bug.
A lot of IT now curates the extensions for the browsers and doesn't allow extensions not on the whitelist and then they basically just only do that work on Chrome and disable Firefox. It's kinda self defeating in the long run imo but that's the problem in the industry.
I've run into a few restaurant sites whose ordering pages just do not work properly (or at all) in Firefox. Also webgl2 performance is unfortunately still much better in Chrome vs Firefox; as an example, FoundryVTT (virtual tabletop software) works fine in Firefox but is a stuttery mess IME (though it has improved slightly in the last few years).
ups.com is one that really infuriates me. It shows 404s for me on Firefox and works perfectly on Chrome.
Kaiser's website works mostly on Firefox. Recently I had to print a "letter" and on Firefox it was blank and printed fine with Chrome.
I don't know if it's still this way, but Google Meet didn't work very well in Firefox, so last year I took all my meetings in Chrome.
These are just what I remember. There are a LOT more.
EDIT: on the UPS thing... it happens when I follow links from gmail in Firefox. Sometimes it wouldn't 404, but I'd see a "..." and it would just stay that way.
EDIT2: for a long time (not anymore), sending Kaiser emails was broken. Hitting enter would warp to the bottom on the page and I'd have to scroll back up to finish typing. They're completely redesigned the website recently and that bug is fixed.
95% of people who use Chrome have no clue what browser they are using.
They got Chrome when it was bundled with every single installer ever for about a decade (which was so prolific and scummy that Microsoft had to make the "default app" picker system more defensive, because Chrome was abusing it more than microsoft apps were).
When you installed Java, you also got Chrome set as your default browser with no interaction.
Or they one click downloaded it from Google.com because of a giant banner saying "You gotta download chrome"
It's insane to me how rarely people on HN seem to actually know the history of this. Everyone who worked in tech support in the 2010s experienced this.
It was an identical strategy that most spyware and adware used at the time.
Not everyone has a well paying tech job. Many have to use their devices until they literally die and many more choose to do so because getting a newer device would mean having to deal with the bullshit of newer software.
I've just 'upgraded' a friend who was using a ThinkPad T410, released Jan 2010. So sixteen years. There was nothing mechanically wrong with it except it was capped at 8GB RAM and chugged along slowly.
Oh yeah, I know "the average person" still thinks Chrome is the best browser (if they even care at all about what browser they use), but I mean in tech circles is where it surprises me...
1. Because it's most popular. Guaranteed support and "monkey see monkey do".
2. The adblocking is preconfigured, and non technical users trying to find the right extensions has a very bad history of unintentional malware. Ad block? Adblock plus? Ublock? Ublock origin? This is a great example of what floors a lot of technical folk who would be "why not just install ublock origin" and fail to understand the "why should I when I can just get Brave one and it works"
If you're anti-Google, use Firefox. It's hypocritical to use the browser they're paying to build, then complain about how they generate revenue to fund it.
After years of using alternative to chrome (Firefox, Chromium, Brave, Opera, Vivaldi, Edge, etc ...) I have stopped fighting the choice of IT for installing and setting Chrome as the default browser on a Mac. I still use Firefox when I can and religiously reroute URLs to it where possible, but this is beating me down and I would rather spend time playing with LLMs rather than continue this struggle.
I use thorium, which also belongs to the empire, so it is not really any different to Brave - but I can use ublock origin still, so that's better. I think we are all in the Google empire here. Praising Brave as alternative, simply does not make a whole lot of sense really.
Firefox is a bit outside of it but it basically got rid of most of its users. When I use firefox, I can not play audio on youtube videos. It works fine with thorium. I tried to convince the firefox developer who said everyone on Linux must use pulseaudio (I don't) but there is no reasoning with Mozilla hackers here. He thinks he knows better than everyone else does. (I could recompile firefox from source, but Mozilla uses mozconfig still: https://www.linuxfromscratch.org/blfs/view/svn/xsoft/firefox... - they are too incompetent to transition into meson or cmake. A failing project, no wonder it lost most of its users. Titanic got nothing on the Firefox team.)
Brave’s owner is a very sketchy dude. With all the news that were happening around brave, all the shitcoin stuff, I wouldn’t be surprised if the browser is mining crypto.
The single affiliated link scandal is enough to not touch that project with a ten foot pole.
I was using Firefox, Vivaldi, Zen, and finally got fed up with various issues that Zen was having, so I switched to Waterfox. I am very happy and the browser is very fast; difference is immense.
To each their own, but I've been using Brave for a long time (5+ years I think?). It was one or 2 clicks to turn off the crypto stuff when I first installed it. It was straightforward and no dark patterns were employed. It has never come back, unlike what Google and Firefox tend to do with their annoying features. It even syncs my preferences to any new browser I add so I only had to do it on one computer once and never worry about it again.
The web's dependency on Chromium engines is deeply concerning, I agree. I used Firefox for a long time. But at this point, IMO Brave is the most pragmatic choice if you want a browser that's not Google but "just works" with the modern web.
Vivaldi just works without annoying crypto and most importantly provides lot of customization, brave feels like barebone browser next to it, hard to understand why someone stays on such basic browser not allowing any customization (tried both and was shocked how bad is Brave many promote here).
I want to use a browser engine that is not developed/owned by Google, so I use Firefox. I also don't want to support Brave's CEO's politics, so I would not use Brave regardless.
My theory is that, since I'm going to do things like banking in my browser, I want one that has a lot of skin in the game. Chrome being backed by Google has trillions of dollars on the line should they ever do anything truly evil. Though this sneaky 4GB download comes close.
It's on-device AI spyware, really. It collects intelligence about the user, summarizes it and sends it to Google, all paid by the user's electricity bill. Deviously clever.
For someone with more knowledge than me: How does this affect other Chromium based browsers?
I did some web searches and see Brave has its own AI thing “Leo” that is intended to preserve privacy. But I don’t think that is on device. Please correct me if I’m wrong.
I use Firefox myself but have family and friends who use various Chromium based browsers.
Brave's "Leo" AI is configurable enough to specify local endpoints for processing, instead of going wherever they want it to go. I've set it up to use my own systems, and it works just fine like that.
If you have a beefy enough device, then yes this can be done on-device.
Chromium mostly does not support this, because it doesn't have the binary blob required to run the inference. However, it does still download the model weights and expose the LanguageModel API, because that part is hooked up.
Packagers might eventually disable that but I tested this behaviour in chromium 148 a few hours ago, and it would download the weights but has trouble running them.
My guess is that this falls under a Google service and the models themselves wouldn't be added to open source Chromium. Even if it were, Chromium forks would likely exclude it like they did for FLoC because of its unpopularity.
The docs say "not yet."[0] My guess is that for Android they probably plan to enable it for high end phones, and for iOS they'll probably just stick to non-API AI features.
Surely there's a googler on here who actually knows whether they are doing this. Anyone actually know or is this post all about Chrome bashing and speculation?
As soon as data starts being exfiltrated to Google (or any Big Tech firm), be sure that governments will demand their copy of the stream too.
The non-disclosure clauses in mass surveillance legislation will ensure the process is opaque to users.
You’ll only find out about it when your door is smashed down and all your devices are seized, because Chrome’s crappy 4GB AI model misinterpreted an innocent photo of your kid in a paddling pool.
These "clicks" are likely identified as fraudulent and dropped by the ad network. So you still pay the cost of downloading and running all the advertising JS and you still get tracked by the ad networks, all for nothing.
You seem more knowledgeable in how browsers and js work than me. Does the below text still mean that AdNausem is downloading and running all the advertising JS?
Here's what's in the link:
>AdNauseam 'clicks' Ads by issuing an HTTP request to the URL to which they lead. In current versions this is done via an XMLHttpRequest (or AJAX request) issued in a background process. This lightweight request signals a 'click' on the server responsible for the Ad, but does so without opening any additional windows or pages on your computer. Further it allows AdNauseam to safely receive and discard the resulting response data, rather than executing it in the browser, thus preventing a range of potential security problems (ransomware, rogue Javascript or Flash code, XSS-attacks, etc.) caused by malfunctioning or malicious Ads.
Basically zero ads are just static images with a link, they're dynamically loaded by JS when you open the page. The JS collects as much tracking data about you as it can, sends that off to the ad network servers which run a live auction to determine who will pay the most to show an ad to you right now, then returns that ad for the JS to display.
AdNauseam not loading the response to the "click" request makes it trivially easy to flag as fraudulent, because a real click would load and run the response.
The same metrics any site uses to identify bot behavior. It's a closely guarded secret because if the attackers knew what metrics they used the attackers would know how to not get caught.
Another reply pointed out that AdNauseam just makes an http request to simulate a "click" and throws away the response. A real click would load and execute the response so it's trivially easy for ad networks to detect AdNauseam "clicks".
Surely this would be illegal? Personal data without consent?
Or is it a case of too big too fail.
Seems like running governments' infrastructure pays off. No regulator will dare to impose a fine that could collapse the company. But this is very much needed.
£100bn fine and confiscation of assets in the given country could be a start.
I too am surprised anyone uses Chrome, but I will admit to feeling similarly surprised by how many people use Brave. The company seems so sketchy to me, and I wonder why people who presumably care about web standards are so willing to use Chromium-based anything too.
This "just keep leaning into the outrage machine" is a terrifyingly effective strategy. I realize lobbying groups have had the "we only need to slip past once" strategy forever, but it feels to me like there's a new level of bare-faced autocratic, anti-social-contract power grab going on in a way that wasn't around 20 years ago.
You see it in our political class. Egregious thing? Nah, try this? Oh you're whining now? Here's some more. Look over here. What about this? Moar gnashing! And now this. Oh you don't like it? Clearly you hate children and freedom and family because this is all and only about protecting the children and saving everyone from rapists, Russians and whatever else the zeitgeist is afraid of.
Why are they so hungry for data? If it's so intelligent, why does it need to learn by imitation so much?
It feels almost like "AI" can't be built without trillions of hours of human work, yet the ownership of the models and the resulting revenue goes only to those in positions of power to exploit that labor instead of the people doing actual work.
I mean to be expected of Google. Even their Google Pay sends data to their servers whenever you use it to make payments, effectively also making it so you can't even use it without service. Apple Pay does not, runs the whole thing on-device, and not only is private, but as a result also enables payments entirely offline.
>Apple Pay does not, runs the whole thing on-device
so when I use the physical card that is also on Apple Pay, and Apple Pay tells me I just made a transaction as if I had used Apple Pay, that is all happening on my device? what online service is my phone using to track my account with Visa or my credit card issuer, and it's polling or push?
You get a notification from Apple Pay when you pay with your physical card? Because I only get a notification from my bank's app whenever I use my physical card. Apple Pay notifications only pop up when using Apple Pay itself.
Maybe it sends the payload after coming back online, but for I can for instance leave with only my galaxy watch 6, which doesn't have esim, and I'm able to make payments as long as I connect it with my phone before leaving the house.
I think the comment's saying that they leave the phone at home, and the watch works by itself as long as it was connected to the phone before leaving the house.
I'm willing to bet that it's just for telemetry, but this kind of stuff just lends credence to the crazies claiming Google wants to create some kind of absurd botnet with people's devices.
It seems to me that adding AI to desktop apps and sending the data back to the mothership for processing is an amazing way to collect data from people who, for the most part, would be completely unaware it's even happening.
Heck, most of them think the Internet is Chrome.
> Heck, most of them think the Internet is Chrome.
In the end Google has achieved something that Microsoft couldn't with Internet Explorer, and won the Browser Wars.
Google managed to aggressively advertise their browser by optional install "offer" within Windows installers of software. And they were aiming exactly at all those who couldn't tell the difference between the web browsers and who were conditioned by more experienced family members, friends etc. to just blindly click "Next Next, Finish". Thus, that was an easy win.
Being here when we had choice between Gecko, Presto, Trident and later WebKit/Blink makes me sad how easily the IT world allowed this nearly 100% monoculture to happen. There are still other browsers but chances that we return to variety and choice of rendering engines are low.
What annoys me the most about this is that we as web developers have been here before with IE and managed to get out of it. Then we immediately forgot that lesson and helped Chrome be the new IE in a new coat by adopting ”modern web API’s” that were in fact proprietary Chrome API’s with low respect for the standards process, until every other browser was forced to accept it as a standard. All those ”Safari/Firefox are lagging behind” or ”best supported in Chrome” wasn’t in our best interest or even, from a standards perspective, particularly true.
At this point Google is almost worse than Microsoft has ever been
Not just allow, they cheered it on.
> adding AI to desktop apps and sending the data back to the mothership for processing is an amazing way to collect data from people
Wasn't that the entire point of Windows Recall as well?
And Siri, for that matter? https://www.scientificamerican.com/article/apple-settles-cla...
Sheesh, I'm starting to notice a pattern...
Every vendor tripped over the third party affiliate human review issue.
While consumers remain surprised by affiliate clauses, the QC problem is considerably different from marketing against those recordings.
The linked article veers into Alexa for the ads part and says, roughly, must not be tin foil hats if everyone believes it – then explains the psychology misleading people in most cases. The "I'm noticing a pattern" thing…
Are there sources where Apple either acknowledges or even settles claims of advertising against secret Siri recordings?
> Are there sources where Apple either acknowledges or even settles claims of advertising against secret Siri recordings?
Yes, advertising and selling data to third parties.
> In January 2025, Apple settled a class action lawsuit brought on behalf of current or former owners or purchasers of a Siri-enabled device, specifically: iPhone, iPad, Apple Watch, MacBook, iMac, HomePod, iPod touch or Apple TV, whose confidential or private communications were allegedly obtained by Apple and/or shared with third parties as a result of an unintended Siri activation. (https://www.nbcnewyork.com/news/apple-settlement-95-million-...)
And also https://gizmodo.com/apple-iphone-france-ads-fine-illegal-dat...
They've also ignored people's explicit requests to not be tracked (https://gizmodo.com/apple-iphone-analytics-tracking-even-whe...)
They've allowed third party contractors to listen to apple customers fucking (and who know what they did with those recordings). Apple's privacy policy says that they'll share (which includes selling) your data with "Apple-affiliated companies" and "our partners" without disclosing any of them.
I've seen claims that they've sold user data to facebook, but I haven't seen any details except the existence of "user-data sharing agreements" between apple and facebook and of course they take millions in cash to hand apple user's data over to google through their search engine.
Apple has been careless about handing data over in response to legal requests (https://www.cnet.com/tech/services-and-software/apple-facebo...) and gives a ton of user data to governments in the US and china. According to Snowden Apple partnered with the NSA as a data provider. Apple denied that. They've also denied working with the US government on DROPOUTJEEP.
They honestly don't have a very good track record when it comes to privacy, they're just quieter and less overtly evil about it than Google, MS, and facebook.
You mean that chrome is an internet explorer?
Something no one else would want - a little colored dot next to HN user names keyed to their generation, so I could quickly tell why none of the commenter mentioned AOL.
What about Prodigy and Compuserve?
More of a navigator for the 'scape of the net.
Great joke, but taking the word literal and not as product name, it makes a lot of sense to describe chrome as a tool to explore the internet.
(Edit: thinking about it, I think generic terms like "Internet Explorer" should not be trademarkable at all, also I just learned, that also Microsoft "stole" the name and had to pay in a settlement..)
No, I think it was meant literally. Like the IT Crowd skit where Jen refers to the Internet Explorer desktop shortcut as "the button for the internet".
https://youtu.be/NUNEZ9-4v_E
According to the current shortage of computing power and electricity, I suspect that what they really want is not your data, but the computing power and electricity from your device.
If users' behaviors can be pre-labeled on their own devices, processed with AI, and then sent back, it might save a significant amount of internal computing costs.
Chrome does/did this already. They've never been of any forensic interest to me but the history file stores clusters of domains visited and keywords searched for. It was pitched as user "Journeys." I think they've deprecated it.
https://chromeunboxed.com/chrome-memories-history-journeys
A 4B model is not very useful for text generation, but useful for classification, so you're probably right.
Yes, it is useful for the next wave of eitch hunting
Witch true/false Commie true/false Possible customer true/false
I called this out when it was announced on here. Supposedly the team lead replied to my comment saying this wouldn't happen. I rolled my eyes but asked will android be able to use those models for ex filtration. No reply. And apparently the original claim was not true either lol.
Maybe I'm misremembering it. Google is awful. My goodness. I hate Android and can't wait to be rid of it. Graphene and it's buddies can't roll it fast enough
Maybe I'm misremembering it.
What you should've done is saved what they said so you could post it directly as evidence. If they're collecting all the data they can, you should naturally also have the right to do the same! I've noticed they're increasingly memory-holing a lot of things that, somewhat coincidentally, are inconvenient truths.
It's in my comment history..sometime last week. I don't really care, I knew this was a surveillance move that's why I called it out. No clue if the anonymous poster was who they claimed to be anyways
Here is the comment: https://news.ycombinator.com/item?id=47930833
They’re all awful.
Does anyone believe a single big tech company isn’t harvesting data en masse from everyone in duplicitous manners?
Like, the best case scenario is that they don’t just blatantly steal your data and instead use dark patterns or inference to take from you without your knowledge.
And then, thanks to the wonderful opinions of the court, the government has full access to said data since you apparently knowingly agreed to giving it to a third party by virtue of the fact that you engaged in any sort of commerce.
It’s why I’m for forcing content being posted on the internet to be non anonymous and tied to a real identity.
The corporations and government already have and abuse all this data. I want the benefit of knowing when someone says “As an American {incredibly divisive shit}” that it’s actually someone in a foreign country sowing chaos for money or political aims.
They won't actually show you who said what though. Twitter trialed that feature then they quickly turned it off after everyone realized half of the maga influencers were russians.
It also kind of stinks because not every mistake should be immortalized and recorded forever. Blackmail and all that. It kind of ruins the internet in a different way.
I want it enforced at a governmental level. I don’t think content consumed by people should be required to be public but if you want to post on a public square that the internet is, I should be able to recognize you as well I could in meatspace.
We get all the negatives of anonymity now with foreign actors, domestic actors, and bots flooding the zone with lies, and not of the benefits since all the corporations and governments can trivially pierce that veil.
I assure you they cannot pierce that veil if one has a modicum of competence.
One uni in my country has been getting bomb threats during the exam period every year for multiple years (a random article says 20 times at least). The whole place gets evacuated each time, nothing is found and nobody is caught.
But people who think they're anonymous because they used a different nick? Yeah, those are idiots, their ISP and the platform knows who they are and anybody can deanonymize them through stylometry.
I don't think surveillance is the solution though. I'd much rather see a network of trust or (second best) anonymous proof of identity.
Any place selling alcohol or cigarettes is able to check if you're 18. They could just as easily check your nationality by looking at your ID and give you a crypto key which can be used to prove that to online platforms without revealing who you are.
But there's no money for big corps in that and most people are not even smart enough to think of it.
Every single packet of information sent on the internet flows through NSA servers at some point. We've known this since Snowden. When I say the government has this information I do not mean that every single bureaucrat at any level of government can access this, but that the government as a whole has access to the information if they feel its worth using. We have limitations on how the government uses such powers but those are currently worth the paper they are written on and being ripped up left and right.
That's defeatist, although certainly what they'd like. The world is not the US.
Even if NSA captured and logged all internet traffic, they'd still only get a fraction of the information within without breaking all the encryption.
And even if they could break the crypto, the ability would only have any power if/when they acted on it. Which in turns reveals the capability to both normal people and other nation states.
The limitations aren't laws, it's the practical consequences.
Of course, having laws with actual teeth helps.
Ok?
So they use it when they feel like and not willy nilly? That doesnt change my perspective. They still have the ability to do it at will.
> when they feel like
No, when they can get away with it. When the pros for them outweigh the cons. And every person affects that calculation in some small way - by their tolerance to injustice, by their outspokenness, by their ability to detect it, by the probability of an employee becoming a whistleblower and even by people's willingness to punish bad actors extra-legally.
> They still have the ability to do it at will.
Yes and your neighbor has the ability to wait on your doorstep until you open and then slit your throat.
He generally doesn't because of internal safeguards (humans have evolved emotions like remorse, though not all of them feel it - see psychopathy/ASPD for details) and external safeguards (the state generally opposes people infringing its monopoly on violence but in the absence of a state, he would still risk being hunted down and tortured/killed as punishment by your relatives and friends).
---
I get what you're saying and I oppose dragnet surveillance but seeing things in pure black and white does not help.
influencers were russians
... or just hiding behind a VPN that exits there?
Could be, but what terrible cover to use.
> Does anyone believe a single big tech company isn’t harvesting data en masse from everyone in duplicitous manners?
TSMC, maybe?
When I said big tech I meant software, not primarily hardware companies. ASML is also not likely to be harvesting our data.
Beyond the easy quip, my point was that "big tech" has taken on a far-too-narrow meaning. I'd bet you didn't really mean "software" either - you were, in effect, referring to the whole amoral (at best) ecosystem of companies which run social media and related web sites and data infrastructure, making their money through addiction, exploitation, and extraction.
Perhaps Cory Doctorow will come up with a better term?
No, I meant companies that have a lot of money and are primarily software. Discord for example is a company that isnt as big as a faang but one of the companies I would have included in "big tech",
The software companies, for whatever reason, have been largely bad actors when it comes to dealing with users data
They don't really count
GrapheneOS is all of the best parts of Android / AOSP.
I'm just not sure how this gives me control of my information, whether I want it sent or not to Google, and if they're retaining it for training or not.
That last question I don't even want to ask because the first two doesn't seem clear.
This could be simply fixed by adding the feature, and defaulting it off, and letting people learn about it and enable it.
> I'm just not sure how this gives me control of my information, whether I want it sent or not to Google, and if they're retaining it for training or not.
That's not the goal. Turning your information into their information is the goal. If training an LLM on some data isn't copyright infringement but instead makes it a brand new non-derivative work, then training an LLM on your personal information arguably means it's no longer your personal information, but instead a legally distinct work.
> adding the feature, and defaulting it off
Nobody gets a promotion for doing that.
Many around the world search in Facebook and only browse via its in-app browser (sometimes zero-rated / free).
I'd argue that even pre-AI the average internet surfer never thinks about all the data the sites they use collect. I'm not even sure if my mom uses any apps that aren't web apps (maybe MS word).
But for it all to go to one place? That's a scary amount of data.
It already does, given how much of the internet is fronted by Cloudflare. They're better positioned than most to know what users are doing.
Compounded with this which is trending on HN, they basically know everything about anyone.
https://sinceyouarrived.world/taken
"Ladies and Gentlemen, may I present to you the Internet."
https://youtu.be/Vywf48Dhyns?si=uOwUSAr1F_QSShfj&t=67
It would be a reasonable deduction for someone who doesn't have the time or interest to understand the internals.
I once had someone ask me why closing the web browser turned off his Internet
The even more frustrating thing here is that after auto-updates everything new [including AI "features"] is turned ON by default.
I do like how Firefox now has a "prevent future AI integrations" checkbox[0], but I just don't believe it anymore (i.e. that it won't magically `uncheck` itself and then enable features I've not requested/authorized).
Which is why I just used an LLM to help me create a local network admin rule to disable the update engine entirely (this SHOULD. NOT. BE. NECESSARY).
[•] <https://www.perplexity.ai/search/b0d3bf5d-7ac7-4d4c-b6c6-32b...>
[0] with a sick darkpattern (for most users to laregly ignore)
[dead]
My belief is that the AI business is all about data collection. The value isn't so much in the quality of the models (that's what enterprise customers and developers pay to get), but in the amount of data that comes "for free" to whoever hosts the models. And then it's worth whoever buys it thinks it is, like insurers or advertisers.
"My belief is the AI business is all about data collection."
The "business" of so-called "tech" companies is all about data collection
https://www.economist.com/leaders/2017/05/06/the-worlds-most...
It’s because they’re all in bed with the Government who wants the same thing. Think of them as an extension of the NSA/CIA and the byproduct of the data collection is being able to sell Ads to make money.
If thats the case why dont they just pay me for it directly.
My guess: people would ask way more for that kind of information. That's why it's a totally immoral business: making people give away something they would never agree to give for free, if they were truly aware of it.
Because the concepts of freedom and security are not universal.
We're way past just collecting for a while now.
it's more like knowledge extraction at this point. younger generations don't build up knowledge any more, everyone else is slowly losing their knowledge by not using it.
Eventually the rug pull comes and knowledge will only be accessible by those who can afford it.
Up until ~2 weeks ago, I believed that at least opting out of data collection would protect me. I no longer do.
Everyone knows that fines paid by companies (instead of the people making the decisions) are considered simply a cost of doing business. A probabilistic tax, if you will.
What finally dawned on me is that given they need more and more data to train bigger and bigger models, at some point the value of using my data for training will exceed the cost of getting caught using it without/against my consent.
There's no escape unless we change the law.
Yeah I was wondering how long it would take for a browser company to do something like this. It lets them scrape data without having to deal with anti-scraping provisions on websites, since now their training data collection gets spread across the entire Chrome userbase and they're able to offload the work of bypassing the Cloudflare captchas or whatever to their end users.
Google doesn't need to that because Google is not affected by any anti-scraping provisions, every website literally begs Google to get scraped.
When you use a coding model running on someone else's computer, you're giving an AI company your proprietary source code and associated documentation, and you're giving free training examples to make a future AI model better equipped to eliminate your job. Valuable data indeed.
Which is why it is odd to see so many companies jumping on the bandwagon, even those that have always been super protective of their oh so valuable proprietary internal code.
Next quarter thinking
Exactly. This is what people don't seem to get. LLMs ("AI") are not a tool for us. They are a tool for the owner class and a replacement for us.
> My belief is that the AI business is all about data collection.
In the short term, maybe. That's what you tell investors.
In the long term, it's about altering, shaping, and even constructing reality: making a new and canonical truth for humanity where the ruling classes are invisible to us and the machine that tells us our history and bedtime stories and how we feel is in every device we carry, until it is everywhere, and it has always been everywhere, and it will always be everywhere.
100% and if you have data other model providers can't 'scrape' (e.g. Google access to Chrome user/usage data) you're in a better position to win.
Yes. It is seriously not a coincidence that all of the ai companies are now offense contractors for the department of war. It's also not a coincidence they want to ban vpns, and force people to verify themselves with IDs, biometrics and their phones for all of their activities. Meanwhile... Bots can run free.
Surveillance capitalism is so stupid.
And right after https://news.ycombinator.com/item?id=48019219 huh
Taken completely by surprise, no one could have predicted this /s
As soon as "don't be evil" became a topic for debate it was over, if you're surprised you haven't been paying attention.
This seems somewhat specious - it's also quite possible that they just altered the wording to make it less verbose. Does anyone have access to the link "Learn more about on-device AI"?
If Chrome starts sending data from the browser back to Google, that's going to be a huge compliance issue. If you work for a company that processes customer data in the browser, you're going to need to ban Chrome.
Chrome has been recording metadata (URLs, timestamps, etc) about your activity since forever, and you can turn this off if you like, see https://myactivity.google.com/myactivity.
They don't record data (POSTs etc).
Nitpicky, but metadata is data and this distinction favors google too much in my opinion
That's on Google apps, that's not on Chrome. That's not Chrome sending your browsing data or content from inside webpages to Google.
"Learn more about on-device AI" links to this: https://support.google.com/chrome/answer/16961953
This seems to be what they're hedging against:
> Some AI features in Chrome do not rely on on-device Generative models, and those features may still run even if the on-device Generative AI models are removed.
it already sends data back to google, the ai stuff, everything that goes in the address bar goes straight to google unless you specifically configure chrome to block it
the on-device ai just offloads some work onto your device
i doubt anyone will be banning chrome, for some reason "it's for ai" is a valid excuse for any amount of sillyness
Yeah, I staunchly refuse to believe an ad company that releases a closed-source browser would violate our privacy. You're probably right that they changed the claim simply because it was too verbose. That's the best and only explanation.
Chrome has been doing that since the beginning.
Good idea.
Not surprising at all. Google went full evil long ago and unless people realise quickly what direction they're going, it will get much worse.
"I am altering the deal. Pray I don't alter it any further."
I mean, you're not paying for it, you've always been the product.
They didn't alter the deal, you just stopped being as naive.
I know that I'm in a bit of a bubble with this one, but I am surprised there is still anyone using Chrome instead of Brave. I get the dependency on Gmail other Google-specific tools, but the built-in ad blocking and Google-free aspects of it made me switch instantly and haven't look back after years.
Brave started off incredibly sketchy and with terrible reputation, for example https://news.ycombinator.com/item?id=18734999
I haven't ever considered it since and I assume many others are in the same boat.
> Brave started off incredibly sketchy
Chrome has stayed incredibly sketchy from the beginning, when Google gained marketshare by sneaking Chrome into the installer for other products that people intentionally downloaded.
Then Chrome did things like "accidentally" uploading your entire browsing history to Google servers when you signed into Gmail.
Now they have declared war on ad blockers, despite the government warning that ad networks are too big a malware vector to ignore.
That's a different kind of sketchy than whatever crypto ad replacement stuff that Brave was accused of doing.
All they did was add their own affiliate link to crypto links that didn't have them. You didn't get tracked from it, and you didn't lose out on anything.
Still sketchy because of the lack of consent, but people act like Brave personally stole money from them.
The other "sketchy crypto stuff" is one of the few actually workable alternatives to funding websites with ads on webpages. Again, Brave took in no money (BAT) that you as an admin / creator would have otherwise had, and they keep it in escrow, they don't claim it.
The only other sketchy thing I can remember is pre-installing a deactivated VPN so that people could pay, push a button and it'd work immediately. Plenty of companies do hacks like that for the sake of UX. Dropbox used to hack macOS its Accessibility permission so people wouldn't have to dive into settings to toggle certain things.
The irony is that Firefox has had their own scandals like surreptitiously installing a hidden ad extension that would advertise for Mr. Robot, but somehow Firefox stans have erased that kerfuffle from their collective memory.
> The irony is that Firefox has had their own scandals like surreptitiously installing a hidden ad extension that would advertise for Mr. Robot, but somehow Firefox stans have erased that kerfuffle from their collective memory.
That isn't what it did.
The only difference is that Google is still doing new sketchy things with Chrome today, two decades later.
The stuff brave actually did was pretty mild.
Its only mild if you are already a well boiled frog.
I've never used brave or any browser that did similar things, so I don't think I'm a boiled frog.
Same here. I don't care how they responded to the backlash, the fact that it happened in the first place was enough for me.
Brave is my default browser for non-sensitive tasks; e.g. most web browsing, GitHub, news, etc. The built-in ad-blocker & tracker blocker alone is worth it. Use chrome for testing. Stock Firefox for anything sensitive.
I'm similar but instead of brave, which I don't trust, prefer Firefox.
I don't trust Firefox either, so I use Zen, which is based on Firefox and also changes the UI.
Zen lost my trust since: https://news.ycombinator.com/item?id=43443494
In my mind, no browser is perfect. However, as far as I can tell that’s not nearly as sketchy as the title implies. It’s for local debugging.
Zen has other issues for me on Ubuntu (eating a ton of resources) which is why I usually use FF. But I put Zen in a different category from Brave and definitely better than Chrome.
I don't understand this choice at all. What do you base your trust on here?
I switched to Firefox when Chrome started messing with the ad blockers. Haven't really had any issues. I prefer developer tools on Chrome but I rarely need to use them anyway.
The trouble is that Mozilla has admitted they can't survive without Google's revenue. You are basically using Google by proxy unless you use a truly independent browser engine of they get blocked by Cloudflare for not having enough fingerprinting tech.
(Ungoogled) Chromium and Firefox are both projects that are open source and readily available. The code is sitting there ready for you to compile. More users = more donations. You can be the change you wanna see.
Neither will accept pull requests that remove the tracking wanted by the corporate overlord.
Mozilla is paid when people search on Google through Firefox. If you're not searching with Google, you're not using Google by proxy.
(Work at Mozilla, but not related to this - this is just public info.)
> You are basically using Google by proxy unless you use a truly independent browser engine
This conclusion doesn't follow your premise. Google has to pay because if Mozilla dies, so does the claim of any real competition on the browser engine market. So everyone agrees Firefox's engine is truly independent. Google pays so Firefox users don't use anything that has to do with Google.
If you think about it, the only real way to not hurt Google is for Firefox to stop existing. Chrome would end up being spun off from Google.
> Chrome would end up being spun off from Google.
You mean, with reasonable administrations, caring for antitrust laws.
I don't agree that you are using Google by proxy when Firefox has more technical independence from Google than Chrome and can be quickly decoupled from the few Google defaults it has, search and safe browsing.
What browser can genuinely claim independence from Google? Chromium browsers are all arguably in the same camp. If FF is implicated, then so are forks like Zen.
Safari is probably the only one?
Safari is funded by Google to the tune of $20B/year https://www.theverge.com/news/769599/google-apple-search-dea...
Ladybug is in the running. We'll have to wait and see where they get to.
Ladybird
Firefox defaults to using DoH with Cloudflare. Less evil than your ISP seeing DNS queries but CF aren't doing it out of the goodness of their heart.
The "less evil" also depends very much on your ISP. At least my ISP has to follow laws that I can theoretically influence.
My ISP also already sees the servers I connect to, so DNS gives them less additional information than it does to buttflare.
I'm just surprised people use Chrome at all. Google has proven over and over they can't be trusted and will exploit you every chance they get.
Because some things only work in Chrome. It's a fact. It's terrible.
We're the frogs being boiled, over the last decade. People sounded the alarms, but they were looked at like they had tin foil on their heads. Now, it's clear they were right.
I'm speaking generally, of course. I use Firefox for all my personal stuff, except for those situations where it doesn't work.
>> Because some things only work in Chrome.
What things? Looks like an urban myth.
Chrome likes to make up new "standards" and then some websites adopt them immediately.
That said, I can only remember two instances of that slightly inconveniencing me in the past, and both times I was inconvenienced by a Google-run website: once upon a time Google Earth refused to work, and once upon a time I couldn't tweak my Google Meet background. Both are no longer the case.
Citation needed. I've seen the opposite--unless there's a very specific niche that can't be otherwise solved, there's huge internal resistance to going it alone.
The biggest counterexample I can think of: WebUSB was critical to Chromebooks supporting external devices, but I can see why Safari might not want it. It has Firefox support at last, though.
Citation of what exactly? That not all browsers implement the same thing at the same time and that some features are Chrome-exclusive because for one reason or another other browsers refuse to implement it?
Is that really something you need a citation on? You sure seem to have come up with an example of your own.
"Chrome likes to make up new standards"
I can think of just one, USB.
Chrome was built on the premise that web standards matter. Remember IE 6?
https://www.theregister.com/software/2026/04/30/mozilla-push...
Remember AMP?
AMP wasn't part of Chrome.
The Prompt API is part of a real W3C standard: https://www.w3.org/2025/03/webmachinelearning-charter.html
It's not even chaired by Google. It's Intel, believe it or not.
I'm aware of a few things, myself:
1) Google properties
1a) Chromecast
2) a few web-based games that were really pushing the envelope on web APIs and didn't bother testing on Firefox
3) WebUSB, commonly used for some things like keyboard customization apps
Which Google properties are Chrome only? I'm not doubting you but the major ones (search, mail, maps, ads) are extremely cross-platform.
In the past there were features that didn't work at all; I used to hit those regularly. Device setup flows, AV features, etc. These days, it's never "this doesn't work on other browsers". It's always "this is worse on other browsers", whether because they don't test it or because they don't care.
YouTube is terrible on Firefox. There was a period where it was usable but got increasingly worse with missed frames, low frame rate. On FastMail and Gmail the expanded search overlay doesn't disappear when you click outside (ESC doesn't work), you often get stuck with it. On YouTube when you stop hovering over the "I like this" etc. on full screen video view, the tooltip doesn't disappear. It's death by a thousand cuts.
It sounds like Firefox needs to implement hardware assisted AV1? I don't see how you can plausibly blame anyone but Mozilla for this.
Firefox has had AV1 HW decode for years.
I wonder if he's using a Radeon gpu, those have had issues like video playback making the entire ui framerate drop. But it happens in steam/electron too, not only firefox. The shader cores aren't used much either, it's just a bug.
A lot of IT now curates the extensions for the browsers and doesn't allow extensions not on the whitelist and then they basically just only do that work on Chrome and disable Firefox. It's kinda self defeating in the long run imo but that's the problem in the industry.
I've run into a few restaurant sites whose ordering pages just do not work properly (or at all) in Firefox. Also webgl2 performance is unfortunately still much better in Chrome vs Firefox; as an example, FoundryVTT (virtual tabletop software) works fine in Firefox but is a stuttery mess IME (though it has improved slightly in the last few years).
I'd bet my bottom dollar those websites still work in Edge, Chromium and Brave. The alternative to Chrome is not Firefox, it's just Not Chrome.
The driver and store signup/portal for doordash returns a 403 forbidden on firefox.
ups.com is one that really infuriates me. It shows 404s for me on Firefox and works perfectly on Chrome.
Kaiser's website works mostly on Firefox. Recently I had to print a "letter" and on Firefox it was blank and printed fine with Chrome.
I don't know if it's still this way, but Google Meet didn't work very well in Firefox, so last year I took all my meetings in Chrome.
These are just what I remember. There are a LOT more.
EDIT: on the UPS thing... it happens when I follow links from gmail in Firefox. Sometimes it wouldn't 404, but I'd see a "..." and it would just stay that way.
EDIT2: for a long time (not anymore), sending Kaiser emails was broken. Hitting enter would warp to the bottom on the page and I'd have to scroll back up to finish typing. They're completely redesigned the website recently and that bug is fixed.
95% of people who use Chrome have no clue what browser they are using.
They got Chrome when it was bundled with every single installer ever for about a decade (which was so prolific and scummy that Microsoft had to make the "default app" picker system more defensive, because Chrome was abusing it more than microsoft apps were).
When you installed Java, you also got Chrome set as your default browser with no interaction.
Or they one click downloaded it from Google.com because of a giant banner saying "You gotta download chrome"
It's insane to me how rarely people on HN seem to actually know the history of this. Everyone who worked in tech support in the 2010s experienced this.
It was an identical strategy that most spyware and adware used at the time.
Why would people still be using a computer from 2010? That might have made sense in 2015, but beggars belief in 2026.
Not everyone has a well paying tech job. Many have to use their devices until they literally die and many more choose to do so because getting a newer device would mean having to deal with the bullshit of newer software.
I have family on fixed incomes. They don't use 15 year old computers. They simply don't last that long.
I've just 'upgraded' a friend who was using a ThinkPad T410, released Jan 2010. So sixteen years. There was nothing mechanically wrong with it except it was capped at 8GB RAM and chugged along slowly.
They're now on a T480, released 2018.
Yeah I was daily-driving a 2012 machine (X230) until 2024, when I then switched to a 2018 ThinkPad which I'm still using.
Oh yeah, I know "the average person" still thinks Chrome is the best browser (if they even care at all about what browser they use), but I mean in tech circles is where it surprises me...
Ok, why Brave though? There's Safari, Chromium, LibreWolf, Ladybird, and plenty of others.
1. Because it's most popular. Guaranteed support and "monkey see monkey do".
2. The adblocking is preconfigured, and non technical users trying to find the right extensions has a very bad history of unintentional malware. Ad block? Adblock plus? Ublock? Ublock origin? This is a great example of what floors a lot of technical folk who would be "why not just install ublock origin" and fail to understand the "why should I when I can just get Brave one and it works"
3. Most people don't use macs
Librewolf meets 2 and 3 (it comes with ublock origin preinstalled), but admittedly fails 1 quite badly.
Not everyone is on Mac. In fact, most people use Windows. So Safari and Ladybird are out of the question, that's two gone.
They mentioned the built-in adblock
Brave is has pre-configured as block that works on everything, also a polished sync experience.
Vivaldi's sync experience is nice as well. Top notch customization too.
Vivaldi is often behind on chromium security patch. In fact they are right now.
I started with Vivaldi. Was unstable in my experience. Constant crashes.
I was very vehement about needing to stay in Chromium — until I tried Zen browser and it turns out I didn’t! (Unless I wanted to watch Prime Video)
If you're anti-Google, use Firefox. It's hypocritical to use the browser they're paying to build, then complain about how they generate revenue to fund it.
After years of using alternative to chrome (Firefox, Chromium, Brave, Opera, Vivaldi, Edge, etc ...) I have stopped fighting the choice of IT for installing and setting Chrome as the default browser on a Mac. I still use Firefox when I can and religiously reroute URLs to it where possible, but this is beating me down and I would rather spend time playing with LLMs rather than continue this struggle.
Well, why would I want to use Brave?
Brave is the Google empire aka chromium.
I use thorium, which also belongs to the empire, so it is not really any different to Brave - but I can use ublock origin still, so that's better. I think we are all in the Google empire here. Praising Brave as alternative, simply does not make a whole lot of sense really.
Firefox is a bit outside of it but it basically got rid of most of its users. When I use firefox, I can not play audio on youtube videos. It works fine with thorium. I tried to convince the firefox developer who said everyone on Linux must use pulseaudio (I don't) but there is no reasoning with Mozilla hackers here. He thinks he knows better than everyone else does. (I could recompile firefox from source, but Mozilla uses mozconfig still: https://www.linuxfromscratch.org/blfs/view/svn/xsoft/firefox... - they are too incompetent to transition into meson or cmake. A failing project, no wonder it lost most of its users. Titanic got nothing on the Firefox team.)
I find Brave's UI uglier than Chrome's.
Unfortunately, there is no way to switch back to the stock Chromium look.
There are other browsers that are free of both Google and sketchy cryptocurrency business!
I use Safari personally. It’s good.
Me too. I don't get the complaints.
Brave’s owner is a very sketchy dude. With all the news that were happening around brave, all the shitcoin stuff, I wouldn’t be surprised if the browser is mining crypto.
The single affiliated link scandal is enough to not touch that project with a ten foot pole.
I was using Firefox, Vivaldi, Zen, and finally got fed up with various issues that Zen was having, so I switched to Waterfox. I am very happy and the browser is very fast; difference is immense.
Waterford -> WaterFox for anyone wondering: https://www.waterfox.com/
Thank you, autocorrect sometimes misfires.
Waterfox is literally owned by an ad company.
Do you have any source on that?
Last I heard they were purchased by System1.
try ironfox sometime. I think its better for privacy.
It seems to be only available for mobile?
You’re definitely in a bubble. Google advertises Chrome on TV. Most users haven’t even heard of Brave.
why would you use brave with annoying crypto and no customization over superior Vivaldi?
To each their own, but I've been using Brave for a long time (5+ years I think?). It was one or 2 clicks to turn off the crypto stuff when I first installed it. It was straightforward and no dark patterns were employed. It has never come back, unlike what Google and Firefox tend to do with their annoying features. It even syncs my preferences to any new browser I add so I only had to do it on one computer once and never worry about it again.
The web's dependency on Chromium engines is deeply concerning, I agree. I used Firefox for a long time. But at this point, IMO Brave is the most pragmatic choice if you want a browser that's not Google but "just works" with the modern web.
Vivaldi just works without annoying crypto and most importantly provides lot of customization, brave feels like barebone browser next to it, hard to understand why someone stays on such basic browser not allowing any customization (tried both and was shocked how bad is Brave many promote here).
Why did I had to come so much down this thread, before seeing a mention of my favorite browser?
[flagged]
This one: http://brave.com/
I don't use their browser but I like their search engine!
"I am surprised there is still anyone using Chrome instead of Brave."
Bubble indeed. No one should use Brave.
I want to use a browser engine that is not developed/owned by Google, so I use Firefox. I also don't want to support Brave's CEO's politics, so I would not use Brave regardless.
+1 for Brave. Been on it for years and it’s fantastic. Strongest security settings without issue.
O no they gave you BAT for visiting websites. Ahhh crypto everyone run!
I'm not familiar with this?
With the BAT aspect? You get tokens for seeing ads. It never really took off but kudos for trying.
Also hilarious that I got downvoted on my main comment but nobody was willing to show themselves.
My theory is that, since I'm going to do things like banking in my browser, I want one that has a lot of skin in the game. Chrome being backed by Google has trillions of dollars on the line should they ever do anything truly evil. Though this sneaky 4GB download comes close.
Google is not liable for your banking.
There's no skin in the game if they do not think they'll be meaningfully punished by government or consumers for their wrongdoings.
And they have trillions riding on milking you for all your data and ad impressions.
Which they seem to think they'll get, regardless of the quality of their web browser. Most people are entrapped by Android anywho.
Edge and Chrome could both be eliminated tomorrow and those trillions would be safe.
You’re the product, not the browser.
I still don't understand why so many people have accepting using an ad company's browser.
The motivation vectors exist here to ensure that, over time, Chrome behaves in ways the end user DOES NOT WANT.
Because they don't know
(that they are an ad company (and they don't know what the implication is)).
When Google did that, did they default the "sending data" feature to off?
Do I even need to ask?
It's on-device AI spyware, really. It collects intelligence about the user, summarizes it and sends it to Google, all paid by the user's electricity bill. Deviously clever.
For someone with more knowledge than me: How does this affect other Chromium based browsers?
I did some web searches and see Brave has its own AI thing “Leo” that is intended to preserve privacy. But I don’t think that is on device. Please correct me if I’m wrong.
I use Firefox myself but have family and friends who use various Chromium based browsers.
Thank you.
Brave's "Leo" AI is configurable enough to specify local endpoints for processing, instead of going wherever they want it to go. I've set it up to use my own systems, and it works just fine like that.
If you have a beefy enough device, then yes this can be done on-device.
Chromium mostly does not support this, because it doesn't have the binary blob required to run the inference. However, it does still download the model weights and expose the LanguageModel API, because that part is hooked up.
https://adsm.dev/posts/prompt-api/#which-browsers-support-th...
Packagers might eventually disable that but I tested this behaviour in chromium 148 a few hours ago, and it would download the weights but has trouble running them.
My guess is that this falls under a Google service and the models themselves wouldn't be added to open source Chromium. Even if it were, Chromium forks would likely exclude it like they did for FLoC because of its unpopularity.
Also, does this affect Chrome for iOS, Android, and iPadOS?
The docs say "not yet."[0] My guess is that for Android they probably plan to enable it for high end phones, and for iOS they'll probably just stick to non-API AI features.
[0]: https://developer.chrome.com/docs/ai/prompt-api#hardware-req...
Google weighs in on Chrome's weights.bin controversy https://www.androidauthority.com/google-chrome-weights-bin-f...
4GB is not "lightweight" nor are local models.
Use anything BUT Chrome or Edge.
I've tried them all but nothing so far beats the UX of Chrome.
Genuinely curious what makes you say that. Haven't used Chrome for a really long time and not once I've missed it.
Surely there's a googler on here who actually knows whether they are doing this. Anyone actually know or is this post all about Chrome bashing and speculation?
Certainly. You think they're going to comment? That itself would be another headline article.
As soon as data starts being exfiltrated to Google (or any Big Tech firm), be sure that governments will demand their copy of the stream too.
The non-disclosure clauses in mass surveillance legislation will ensure the process is opaque to users.
You’ll only find out about it when your door is smashed down and all your devices are seized, because Chrome’s crappy 4GB AI model misinterpreted an innocent photo of your kid in a paddling pool.
Gemini is also the only major provider where you can't opt out of using your data for training without disabling chat history.
Since the thread evolved into browser comparisons, I'd like to endorse a better uBlock ('s fork) - AdNausem.
It doesn't block ads. It clicks them first, and then blocks them.
I don't want websites to loose revenue because of my adnlocker. I want them to make extra money because of it!
I'm not affiliated, but would like the project to get more followers. This can stop ads once and for all.
These "clicks" are likely identified as fraudulent and dropped by the ad network. So you still pay the cost of downloading and running all the advertising JS and you still get tracked by the ad networks, all for nothing.
https://github.com/dhowe/AdNauseam/wiki/FAQ#how-does-adnause...
You seem more knowledgeable in how browsers and js work than me. Does the below text still mean that AdNausem is downloading and running all the advertising JS?
Here's what's in the link: >AdNauseam 'clicks' Ads by issuing an HTTP request to the URL to which they lead. In current versions this is done via an XMLHttpRequest (or AJAX request) issued in a background process. This lightweight request signals a 'click' on the server responsible for the Ad, but does so without opening any additional windows or pages on your computer. Further it allows AdNauseam to safely receive and discard the resulting response data, rather than executing it in the browser, thus preventing a range of potential security problems (ransomware, rogue Javascript or Flash code, XSS-attacks, etc.) caused by malfunctioning or malicious Ads.
Basically zero ads are just static images with a link, they're dynamically loaded by JS when you open the page. The JS collects as much tracking data about you as it can, sends that off to the ad network servers which run a live auction to determine who will pay the most to show an ad to you right now, then returns that ad for the JS to display.
AdNauseam not loading the response to the "click" request makes it trivially easy to flag as fraudulent, because a real click would load and run the response.
What metrics does the ad network use to identify the clicks as "fraudulent"?
The same metrics any site uses to identify bot behavior. It's a closely guarded secret because if the attackers knew what metrics they used the attackers would know how to not get caught.
Another reply pointed out that AdNauseam just makes an http request to simulate a "click" and throws away the response. A real click would load and execute the response so it's trivially easy for ad networks to detect AdNauseam "clicks".
How will it stop ads if it rewards them with money?
It makes them burn money with no effect. Doesn't work every time, but still sends a message.
It rewards Google with the advertiser's money, and the advertisers don't like paying for extremely low conversion rates.
Because it could eventually be detected as click-fraud, and ad networks hate paying out for click-fraud.
You question is the answer to your query
Surely this would be illegal? Personal data without consent?
Or is it a case of too big too fail.
Seems like running governments' infrastructure pays off. No regulator will dare to impose a fine that could collapse the company. But this is very much needed.
£100bn fine and confiscation of assets in the given country could be a start.
> Surely this would be illegal? Personal data without consent?
I'm not trying to be mean here, but have you been frozen in ice for the last 20 years? This is effectively the tech industry's raison d'etre.
I too am surprised anyone uses Chrome, but I will admit to feeling similarly surprised by how many people use Brave. The company seems so sketchy to me, and I wonder why people who presumably care about web standards are so willing to use Chromium-based anything too.
Temporary escape hatch: https://github.com/ungoogled-software/ungoogled-chromium
I wonder if this is in response to the chrome incognito lawsuit.
Google is a disease. We should stop spreading it.
"on-device" is doing a lot of heavy lifting when the device is a thin client to Google's servers wearing a trench coat.
Has anyone found a browser with comparably good dev tools to Chrome?
You could try Helium (https://helium.computer/), it's a de-googled chrome and has the same devtools.
nice, thank you
Replay.io has a browser that does time travel debugging, which is really really neat.
What makes then so good? I always try them and then go back to Firefox.
Would ad blockers stop this traffic from being sent
This "just keep leaning into the outrage machine" is a terrifyingly effective strategy. I realize lobbying groups have had the "we only need to slip past once" strategy forever, but it feels to me like there's a new level of bare-faced autocratic, anti-social-contract power grab going on in a way that wasn't around 20 years ago.
You see it in our political class. Egregious thing? Nah, try this? Oh you're whining now? Here's some more. Look over here. What about this? Moar gnashing! And now this. Oh you don't like it? Clearly you hate children and freedom and family because this is all and only about protecting the children and saving everyone from rapists, Russians and whatever else the zeitgeist is afraid of.
as if they didn’t have enough data already, good choice to lose any remaining trust from the public over this
Google and Meta always phone home….
Why are they so hungry for data? If it's so intelligent, why does it need to learn by imitation so much?
It feels almost like "AI" can't be built without trillions of hours of human work, yet the ownership of the models and the resulting revenue goes only to those in positions of power to exploit that labor instead of the people doing actual work.
I mean to be expected of Google. Even their Google Pay sends data to their servers whenever you use it to make payments, effectively also making it so you can't even use it without service. Apple Pay does not, runs the whole thing on-device, and not only is private, but as a result also enables payments entirely offline.
> Apple Pay does not, runs the whole thing on-device, and not only is private, but as a result also enables payments entirely offline.
Apple Pay still does send a lot of telemetry about your payments though. https://duti.dev/randoms/wip-location-services/
>Apple Pay does not, runs the whole thing on-device
so when I use the physical card that is also on Apple Pay, and Apple Pay tells me I just made a transaction as if I had used Apple Pay, that is all happening on my device? what online service is my phone using to track my account with Visa or my credit card issuer, and it's polling or push?
You get a notification from Apple Pay when you pay with your physical card? Because I only get a notification from my bank's app whenever I use my physical card. Apple Pay notifications only pop up when using Apple Pay itself.
> You get a notification from Apple Pay when you pay with your physical card?
I do. Which is sometimes annoying if somebody else is looking at my screen.
Maybe it sends the payload after coming back online, but for I can for instance leave with only my galaxy watch 6, which doesn't have esim, and I'm able to make payments as long as I connect it with my phone before leaving the house.
Google Pay works for a limited amount of uses in offline mode.
https://9to5google.com/2023/12/20/google-wallet-without-inte...
If your phone doesn't have connection does it still work on your galaxy watch? Or if you leave the phone behind?
I think the comment's saying that they leave the phone at home, and the watch works by itself as long as it was connected to the phone before leaving the house.
Wow...that seriously may change my long standing anti-Mac disdain to pro-Mac advocacy, very interesting, even Gemini confirmed what you're saying.
I'm willing to bet that it's just for telemetry, but this kind of stuff just lends credence to the crazies claiming Google wants to create some kind of absurd botnet with people's devices.
Al or AI?
It's Google. It's AIs
Just today Google launched it's health app on Android and promised to not use people's data to sell them ads.
I called that bullshit, guess this article is just proving my point.
What we learn: we can not trust Google.
Everything made by Google is a liability.
Doesn't look like that has been or will ever be (generally) learned.
You're just now learning this? There are whole books about it (check out "The Age of Surveillance Capitalism" by Shoshana Zuboff)
They're probably doing some degenerate form of [1].
[1] https://en.wikipedia.org/wiki/Parasitic_computing