ragequittah

Notable that when it couldn't count the "r"s in strawberry the LLMs were already solving problems most of the people laughing at it for not being able to count "r"s wouldn't be able to come close to figuring out with weeks of research. People have had a myopic view of this technology for a long time. Many still do.

I always try to parse this "Mozilla is doing bad" as I've been using Firefox for years and other than features I want (vertical tabs, firefox relay, and a few other minor things) I couldn't tell you the difference of Firefox now vs 5 years ago vs 10 years ago. And the "shooting yourself in the foot" things talked about are always something that pretty much all other browser companies do but 10 times worse.

I think the last major scandal was AI additions (cuz who is using AI these days? What an awful idea!) that you can now turn off with one click? Personally never saw the AI features presented to me once. Then the one before that I can remember was the Mr. Robot thing that was so long ago it's hard to remember what even happened. I think they installed a add-on that wasn't enabled and did absolutely nothing unless you went looking for it and specifically wanted it to do something? And then they removed it very soon after when nerds overreacted so hard that we're still talking about it 10+ years later?

Chrome: inability to block ads, new "cookie free" (nope they didn't get rid of cookies) ad tracking, monitoring people in incognito mode, the list would goes on for a long while. Brave: Crypto nonsense, all kinds of advertising for their crap everywhere unless you go through many hoops to turn it off. Firefox is one of the few pieces of software I can say hasn't enshitified itself to the point of using something else.

I feel like if the doomers are correct and AGI happens research can't say anything about it. We won't be able to predict the outcome of something more intelligent than us or that can disrupt society so thoroughly. It's already causing major problems.

I follow AI quite closely and I have yet to see any hostility towards artists or mathematicians from anybody worth talking about (maybe some forum dwellers?). Nor have I ever seen any marketing or implications from big AI companies that says creative work or math is in any way useless or obsolete. The fact that AI can solve long unsolved math problems isn't hostility towards mathematicians.

I'm very envious of your workplace if the stuff before AI was somehow well written. If I had AI 10 years ago I would've been mostly using it to interpret most of the corporate speak e-mails and poorly written technical briefs.

I just wonder how many people smugly cry 'slop' and then immediately go ask chatgpt about something they want to know about. I suspect it's a surprisinly large amount.

More like an alarm that deters (though not always) the robbers. You can still very easily sideload apps.

What about an entire family of people who have no interest in learning technology they just want the shady ad-free youtube on their phone? I find it perplexing you've never met people like this. I'd wager it's more common than not in our world.

But that's just a completely different discussion that we could've had a long time ago when they introduced it as a on/off security setting ~15-20 years ago too (and I bet people did). IMO nothing has changed except I bet more people will take going into developer options, being sternly warned, then having to wait 24 hours as a sign this is something maybe they shouldn't be doing. And I'd argue if you come to that conclusion and don't look into it further you're probably right,

Given that google is now forced to have competing app stores due to Epic v. Google and the ninth circuit courts went into long and arduous detail in that anti-trust case about sideloading restrictions I don't think they have a leg to stand on if they get rid of sideloading completely. Oddly Apple continues to get away with it no problem.

You do understand we've broadly come to the conclusion that this isn't how society should work though, correct? It's why there's warning labels on everything, liability for not warning people for even the dumbest of things they might do with your product or in your establishment, extremely strict laws for things like driving a car, flying an airplane, or using any kind of heavy machinery, responsibility of servers to cut people off if they've had too much alcohol, the list goes on. I find intelligent people often have this mindset that all "real adults" are just as responsible as they are. It's just not the case.

And I generally wouldn't care "have at it" if their phone/computer being compromised only harmed them. But a bulk amount of compromised devices harms us all.

"Think of how stupid the average person is and then realize half of them are stupider than that!" -George Carlin

>so ultimately I think some default level of "guardrail" is appropriate, but it should have a not too burdensome escape hatch

Which is really exactly what google is implementing here and why I'm so perplexed so many people are upset. You or I or smart parents (my mother has never used a computer in her life and wouldn't want to) will have absolutely no issue sideloading apps after the proposed changes.

Have you ever tried to educate a 70+ year old person on this kind of stuff? Their eyes gloss over and they don't care. I've told my mother not to install random apps on her phone several times yet she still does it. I could probably send her a phishing e-mail right now that 'looks like' it's from me (so long as my name is in the from: field that's enough) and she'd follow the instructions to install a malicious apk too.

"Everyone should just be smarter/more educated" is what I used to think as well. I now fully understand some people have no desire to be educated in any way and it's a net good for society to not have a large % of devices compromised.

Surely you see how installing unvetted software on a device that holds your entire life, bank info, photos, etc. probably shouldn't be part of the "happy path" for most people. You shouldn't even install software you haven't fully vetted from the google play store. This should be drilled into non-technical people.

Those of us who want to can easily bypass it. My mom who doesn't know what she's doing and gets a phishing email with a fancy apk attached should probably have a hard time using it.

I'm not sure I understand this argument. If I go to the library every day for 10 years and learn everything there is to know about subject x I shouldn't be able to sell my skills to the world about it later because I didn't give the creators of the books I read any money?

Arguably LLM companies could have made large-scale deals with libraries and got the exact same knowledge (much, much more slowly). I wonder if people would have the same issues then? My guess is probably. Goes back to the meme that if libraries were proposed today there's no way they would ever be allowed.

My bet: it's less than a 2% loss in customers that are just very loud on the internet.

Alternatively: if you're wasting a bunch of your time on the minor details how do you have enough time for the major ones? I'd much rather my restaurant owner worry about sourcing ingredients, keeping the kitchen clean, improving the taste of the food than anything else.

This is something I've thought about as people complain about things like restaurant photos being AI and such. You know what kind of customer this pushes away? The kind that complain about minor details that don't really matter. Might be a net good for many businesses.

Major companies are using these LLMs on their already hardened software and finding countless thousands of security flaws. Is that all theater too? If not then it's very easy to see how these things are dangerous.

I made an AI video of a picture of me when I was about 2. Had a huge smile on my face in the photo so as a lark I made a 5 second video of me laughing. The look in my mom's eyes when I showed her and the way she grabbed the phone from me to look closer was such a good moment.

It's not an alternate reality. I was laughing at the time almost certainly. New ways of seeing old memories really needn't be depressing at all. It's so strange how so many people seem to only see the bad in a technology that can be used in countless different ways.

Windows 11 LTS is a dream to use. Unfortunately I need 25H2 for dumb gaming reasons so I can't use it. It proves Microsoft can make windows properly they just refuse.

When I opened up the old notepad and my add/remove programs was 3 entries long I couldn't believe it.

The problem is they really are the best TVs. I have one sitting next to me that thankfully hasn't been connected to the internet in 5 years. Take the subsidies they give you for being able to spy on you then don't let them spy on you is the better move.

These companies are going too far too fast like we always do as humans, and I agree the negatives aren't going to be pleasant. But there are industries and products out there that we all use currently that have way worse externalities that we just shove to the side in our minds because they've been around since before we could yell on twitter about them. The beef industry is a nutty thing to exist in its current form. Oil and plastics what the hell are we even doing? My guess is if you put your head down and did the research AI doesn't even rank in the top 10 of problematic things that people use daily.

Air travel, everyone having a nice green super watered lawn / golf courses, air conditioning in every home, fast fashion. A single person using AI for their work doesn't even register on the list of offences if you compare it to these things. Driving an ICE car every day for more than 30 minutes is wild. Hell it probably doesn't even use as much power as playing a video game on a good graphics card for 2 hours.

The thing I've resigned myself to is very few people will change anything at all for the greater good of society that gives them even a minor inconvenience. See: COVID restrictions. My honest take is we're absolutely doomed as a species unless we come up with novel solutions to the problems we're facing. And guess what AI is showing itself to be extremely good at in various fields [1][2][3]? It's a pipe-dream hope, but the hope is there. And AI when compared to the decades (sometimes centuries) of oil, beef, coal, whatever else won't even be a pixel on the screen of reasons why climate change is happening for some time yet. And it will highly incentivize mass production/adoption of the cheapest power sources which also happens to be renewables. I understand it doesn't look like that right now as people like Elon make their disgusting data centers, but over time I'd bet on it only because the financial incentive is there.

The discussion is far more nuanced than "AI bad". I understand the horrific aspects of it, but there is plenty of good to be found there too. Accessibility, translation, help with learning disabilities. The list of good outcomes goes on for a long while. It's a powerful tool that can be used for good or ill. In my opinion we need more people with the principles you speak of to be using powerful tools for the benefit of those principles.

[1] https://news.mit.edu/2025/using-generative-ai-researchers-de...

[2] https://www.nature.com/articles/s41586-021-03819-2

[3] https://www.ipd.uw.edu/2025/12/rfdiffusion2-efficient-enzyme...

>They are poor tools that can't do the things they are claimed to be able to do

This is the one I don't understand the most out of all the gripes. I find it genuinely unsettling how much I've done over the past couple of days (and with little to no mistakes or bugs in the code that I can find) by just broadly asking Astra to do it. I understand why you'd feel the technology sucks 2 years ago, but these days the models are becoming way too good at what they do.

I agree that if it lives up to the hype we're in for major societal upheaval. Existential crisis is a very real part of looking at this tech for what it is. But I also don't see how not using it stops that. It'd be like not ever taking an airplane to prevent climate change and wondering why it didn't work.

The fun is in being able to finish projects you've had rattling around in your head for years but never had time to do. Or researching and discovering things that you'd never bother with previously because the internet has become an impossible to navigate cesspool. Or learning things in a way that is completely customized to your speed and level of knowledge.

I have to hold 2 thoughts in your head at the same time: the technology is empowering at an individual level while also being highly disruptive at a societal level. We're not going to be able to put the toothpaste back in the tube at this point.

There was a whole gaggle of people who refused to get it for that and a few other "user friendliness" reasons. And I completely understand it.

>Across all groups of users, taking too long to download pages (i.e. "speed") is the most commonly experienced problem with the web (64.8% of respondents).

>Online retailers especially should take note: 53% of respondents reported that they had left a web site while searching for product information simply because the site was too slow.[1]

It's strange to think about now but roughly half of people in the US didn't have internet in the year 2000. And half of people that had the internet weren't making any purchases. People didn't trust or see much use for the tech back then and it was really hard to convince them.

[1]https://sites.cc.gatech.edu/gvu/user_surveys/survey-1998-04/...

You can also find something uninteresting without brow-beating people about it. Goes both ways. Just giving my experience with it, take it or leave it. The comment I was replying to seemed to wonder why people didn't understand non-engagement with the tech, I gave my reasoning. The cost of engaging is also so slim (and it's freaking everywhere) that it has to be an almost religious aversion to not, which I find quite perplexing in itself.

Personally I generally agree with guardrails for regular consumers. The kind of people who don't want to do things like go into developer options and enable "unsafe" options usually should be protected against installing a random .apk on a device that contains their bank info and whatever else.

I just find it such a fascinating technology. It's finally fun to be a nerd again! For me it's the same feeling as when I was on the internet in the early '90s and was trying to tell people how useful it would be. I don't hold a grudge against anybody who held out until the iPhone came around or whatever it was that got them to finally realize the usefulness, but I still question why it was so hard to convince so many people when the benefits were just plainly obvious.

If you want to avoid the technology go nuts. I just hope you're doing so knowing how important it will be to the future. It's also almost satire how outdated the assumptions of many people who don't engage are. "It mostly lies to you anyway" or "It can't even count the number of letters in a word" are such silly things to say about the technology these days. As is the typical "but it uses so much water!" as the person orders a hamburger.

My gripes mostly come with people who don't engage with it and then spout memes/misinformation (ironic) about AI as if they're experts. Usually akin to "That internet stuff is garbage, pictures take 2 minutes to load!" after everyone and their dog already has broadband.

Humans adapt to change oddly fast. If you saw that video in 2022 you'd be 100% positive you were watching a sci-fi movie.

You can still install your own apps and in my case your own operating system (graphene). The 24 hour wait to activate sideloads indefinitely seems like a non-issue to me.

https://support.google.com/android/answer/17588095

You have to bake the 'lazy dev'/'keep it simple stupid' mentality into your AGENTS.md and / or the skills you're using to design things. It will take things too literally sometimes so you also have to make sure you're being accurate. Best way I've found to use it is make it ask you clarifying questions about what you're trying to build and have it help design the shape of the thing. Then it writes the instructions in a format it understands.

I've had Claude do the same thing where it goes off and spends 100% of my tokens on 3 functions and an ungodly amount of tests / scaffolding that do almost nothing when I gave it an underdeveloped idea.