chromacity

For me, practical knowledge comes from trying to figure things out. The more polished and "ELI5" the material is, the less I retain. I've played with quite a few LLM tools that promised to help me "understand anything", but I don't think they help with intuition all that much. For what it's worth, it's not an LLM-specific problem. I like YouTube content like 3Blue1Brown, but I don't think that I retained anything useful from any of it.

I don't question that LLMs are useful for answering questions about codebases, but this is closer to "turn a codebase into a curriculum", and... does that actually work?

I think what you're attributing to frugality might be a more a matter of age? Many older folks are just wary of change.

I'm not that old, but every time I upgrade my PC or phone, some of my workflows break and I need to pointlessly re-learn things I'd rather not re-learn. UI buttons get moved around, icons change, some settings are removed and others are added... this was exciting the first ten or twenty times, but it's just tiring now.

Basically, I'm at this stage in life where my reaction to systemd wasn't "oh wow, this is progress" but "ugh, I need to learn how to start, stop, or modify services again". In another ten years, I'll probably just say "no, I'm not doing this again, just let me use my old computer for as long as possible".

Funny, I went the other way round in the Bay Area. PG&E bills were so high so it was the choice of putting on a jacket or paying $1k extra over the winter months. And my reasoning was "I can afford a jacket".

Most people in the US are pulled into living on credit straight out of school. You get a student loan, then a car loan, then a credit card, then a mortgage. You finance vacations, appliances, kitchen remodels, smartphones - mostly to keep up with friends and coworkers who finance their lifestyles too. A lot of people are in non-stop debt from the age of 18 to 55, if not longer. By most estimates, only about 10-20% of US households are debt-free.

Spending and getting into debt are useful tools. But I don't have any friends in tech who need to be told "hey dude, you should be spending more". I have quite a few friends who would be better off spending less.

Note that the title is completely misleading. When they say "creativity", they mean it in the sense of a "marketing creative". This is about marketing people evaluating AI-generated product shots, which are hardly the pinnacle of human creativity.

I remember hearing the same arguments in the early 2010s, when the "3D printing revolution" was just around the corner. Why would anyone buy anything anymore if you can download a model and print it in the privacy of your home? And make it infinitely customizable?

The whole point of having a civilization is that most things in life can be made someone else's problem and you can focus on doing one thing well. If I'm a dentist or if I run a muffler shop, there are only so many hours in a day, so I'd probably rather pay a SaaS vendor than learn vibecoding and then be stuck supervising a weird, high-maintenance underling that may or may not build me the app with the features I need (and that I might not be able to articulate clearly). There are exceptions, but they're just that, exceptions. If a vendor is reasonable and makes a competent product, I'll gladly pay.

The same goes for open source... even if an LLM could reliably create a brand new operating system from scratch, would I really want it to? I don't want to maintain an OS. I don't want to be in charge of someone who maintains an OS. I don't necessarily trust myself to have a coherent vision for an OS in the first place!

I don't see what the issue is, my agent is already running as root.

> There's an incredibly serious lack of education with how LLMs & carb-counting works

Oh! Do the vendors offer trainings to make sure the users understand how LLMs work? If not, surely, the LLM itself is trained to know its limitations and politely decline in situations like that?...

The #1 use case for this tech is "here's a problem I don't feel like solving, let's have a computer do magic". It's how it's advertised on TV, it's how it promoted in the software I already use. Food preparation? Travel planning? Shopping? Tutoring your children? You can do anything now!

I just talked to a realtor who will make a killing on a real estate transaction. Instead of offering human insights, they sent me "AI reviews" of several properties. The AI has never been to any of these properties and has no idea how they actually look like. But I guess it's how we operate now as a society.

If you go to eBay, every other listing description for used items is AI-generated. This is an official platform feature for sellers. The AI doesn't know the condition of the item or what's included or missing. Doesn't matter, it's magic. It's AGI, it will figure it out.

Most of the uses of AI I encounter as a consumer are like that, and the companies selling this tech are 100% complicit.

> The revenue from a few ads on the free tier in exchange for limited queries to GPT-5.3 is negligible

So why chase this negligible revenue?

> We need better critics of the industry.

There's plenty, but they don't have enough material to post once a week. And if you don't post once a week, you don't end up on HN once a month. As simple as that. Looking at the blogs that show up on HN regularly, the usual hit rate is 10-25%.

That's essentially how you become an online pundit. The internet rewards provocative takes. If you have a tendency to doubt yourself and revise your views, then (a) your views become less provocative and thus less likely to translate into click-worthy headlines; (b) you end up biting your tongue or saying "I don't know" often enough that is becomes impossible to keep up with the requisite weekly publication schedule.

Which is to say, it's easy to scapegoat this guy, but I think his approach is not any different from other "opinion piece" bloggers that we all tend to reshare.

Third evidently AI-generated "AI is bad" story in a day. I'm gonna lose it...

There's always the Juiceros of the world. More seriously, every software company of note has some hardware aspirations and hires some number of EEs, machinists, material scientists, etc. Not as many as SWEs, but if you can get your foot in the door, it's probably nice.

> As long as a judge issued the warrant for geofence data, I see less wrong with it. It passed judicial scrutiny, AND can itself be challenged.

The cops say "someone committed a crime in this area, we need to find the perp". They can pretty much say this for any part of the town at any given time. A judge signs off on the warrant, because why wouldn't they? You don't get to challenge anything: no one is going to tell you "hey, your phone was in that area, come to the courthouse and make your case if you think the police shouldn't be given that info".

It's striking how many of these "product safety" cases are decided in the court of public opinion, independent of actual scientific merit. The case of DDT was pretty interesting. More recently, we have microplastics - no one has really shown they're dangerous to humans, but there's enough hand-waving that "everyone knows" they're killing us. And aspartame, etc...

Glyphosate is probably the safest of the things people spray their lawns with. I don't think we should - the worst you get on a typical suburban lawn if you mow but don't spray are dandelions and clover - but it's probably not giving you cancer. As for food... again, there are far worse, more persistent pesticides that escape this kind of scrutiny.

> It's not a great definition but it's also not a terrible one either. For an AI system to be able to do all or even most of the jobs in an economy

That's not the definition they have been using. The definition was "$100B in profits". That's less than the net income of Microsoft. It would be an interesting milestone, but certainly not "most of the jobs in an economy".

> I was surprised not to see any discussion on whether the author used AI to help write this post.

It is definitely AI-written, far beyond "AI assisted". This is a shower thought turned into a needlessly long machine-generated essay that doesn't say anything a chatbot wouldn't say if you said "hey ChatGPT, write me a thought-provoking essay on <x> for HN".

I made a comment about it, along with several other folks, but the thing is... we get these AI-written "AI is bad" / "AI is great" articles multiple times a day. Debating them doesn't scale, but neither does complaining about them, and especially not complaining in a thoughtful way. Most people on HN are content to argue with a machine.

Aaand it's the second "AI is bad" story on the front page today that's evidently generated by AI.

Everyone has an identity. We have people with near-religious beliefs about AI, people who cram functional programming where it doesn't belong, etc. Our hobby projects are often a consequence of these identities and make no sense otherwise. A guy who builds a web server on a Z80 CPU is doing something fundamentally pretty stupid, but we like it, right?

So, how does a Z80 webserver differ from a PCB made out of clay? Why does this particular project need to have the right kind of ideology underpinning it before we can enjoy it?

If we're uncomfortable or "have questions" because someone brings up feminism as a justification for their geeky hobby... that's on us.

I'm not sure if this is what you mean by "distribution", but I think Instagram won by being content-consumption-centric. People who used Flickr were mostly other photographers. People on Instagram are mostly your customers: people planning their own weddings, moms with newborn children, etc. It's basically a marketing platform, so it's a no-brainer for current or aspiring pros.

It sucks for passionate hobbyists, but photography is first and foremost an industry.

I just can't get over the incredible irony of so many of these "AI is bad for you, mmkay" articles being LLM-generated.

If the author sincerely believes the thesis that AI makes you vulnerable / dumb, they are either incredibly hypocritical. But more likely, they're just cynical and trying to get traffic to their website. And you're not getting back the time you spent reading this and arguing with it.

At the peak of its popularity, Flickr was an interesting glimpse of the coming age of algorithmic homogeneity. In the mid-2010s, most of their top photos looked basically the same: heavily shopped, oversaturated HDR landscapes.

I stopped using Flickr around the time they started flirting with bait-and-switch strategies - "we'll hide / delete your old photos unless you pay" - so maybe things have gotten better... although I see that artificially-looking landscapes still dominate their "trending" page (https://www.flickr.com/photos/tags).

Anyway, my general takeaway is that things are more interesting on photo sites where engagement isn't driven primarily by a global popularity ranking. You just come across thought-provoking work more often.

If it's an existential threat to humanity, and if OpenAI is valued at nearly $1T, why set the bounty at a measly $25k? The going rate for an iPhone zero-day is six to seven figures. Some companies will pay you more than $25k for a website XSS.

Because this is not a serious effort to address a serious risk. It's a PR stunt, the bounty is for a simple jailbreak and not a bioweapon, and they don't necessarily want to spend a lot of money or get people really invested in breaking their safety filters.

Because it can't and it's a publicity stunt. It achieves three goals:

1) Underscores to the general public that the models are amazingly powerful and if you're not using them, your competitors will out-innovate you,

2) Sends the message to regulators that they don't need to do anything because the companies are diligent to prevent harm,

3) Sends the message to regulators that they sure should be regulating "open-source" models, because these hippies are not doing rigorous safety testing.

Both Anthropic and OpenAI have been playing that game for years.

It's solvent-based, so it won't set well in thick layers and it will shrink significantly as the solvent evaporates. You can do thick layers with solvent-free thermoset resins such as epoxy, but epoxy will yellow over time.

It's a relatively soft plastic and I don't think you can realistically build a uniform, good-looking layer that's 1/8" thick, if that's what you mean. If you need that thickness, high hardness, and nice appearance, I think your best bet is just a sheet of glass or acrylic on top.

It can be used as protective varnish, but that would be a very thin layer, probably 0.1 mm or something like that.

I think that gen AI has profound negative externalities. I also think that all the other uses of this tech aside, natural language comprehension alone is cool and useful enough that I don't see us ever going back.

So I guess my question for the author is what they're trying to achieve in this essay. Some of the big players in tech may have beliefs that align with some of the tenets of fascism. But the label is so well-worn that it's meaningless. Buying books on Amazon is fascism, writing on Medium or Substack is fascism, having pets is fascism, etc. So, this day and age, not much is accomplished by uttering the word.

With this in mind, what's the prescription the author actually has to offer? That's where it really gets dicey, because the only takeaway from this writing seems to be "geez, someone ought to set some datacenters on fire".

Dunno. The internet was definitely smaller, but it was also largely uncorrupted, so you could literally just email a random university professor or an industry expert and get answers to dumb questions.

And today, if you want to learn something the right way, you probably still should buy a book (or, I guess, pirate an ebook). I don't think you can really learn much from YouTube influencers and the like.

I think this is a weird side effect of how we portray evil corporations in fiction and in journalism. We imagine that everyone working there is a moustache-twirling villain. And then we get a job at Meta or Flock or Palantir, look around, and don't see any moustache-twirling villains. There's no one saying "ha ha, we should hurt people just for fun". So, it must be that we're the good guys.

Even if some of the outcomes seem reprehensible, it's not really evil because we're good people. We do it in a responsible and caring way. We're truly sorry that your grandma is now hooked up on endless AI-generated slop, but shouldn't the media be talking about all the other grandmas whose lives are enriched by our AI? We have strict safety rules for the types of cryptocurrency ads that can target the elderly, too.

Using a 1960s book as a benchmark feels weird to me. I'd expect books to be more expensive when they come out and less expensive when they're the fiftieth low-cost reprint 60 years later. Sure, it's a classic, but it's hardly a "must-have". At best, it's something you need to read for school, although many school districts have dropped it from their lists.

Having said that, I think the complaints about book prices are mostly an excuse for preferring to spend time on social media or download pirated books for free.

Leaving aside the question of whether they're priced "correctly", books are cheaper than a Doordash meal or a computer game we buy and never finish. Would the average person really read more books if they were $4.99 instead of $29.95?