lich_king

Since the discussion here focuses mostly on child prodigies in general, I'd note that there are many studies showing that the usual outcomes for gifted kids are not all that great.

I think the issue is that it's just harder to fit in. I remember being way ahead in some classes in middle school, and I actually ending up drawing the ire of some teachers when I had answers to every question (let alone corrected them). I eventually learned to disengage and just look out the window. But if you develop that attitude, you never learn how to cram in knowledge for tests, which actually increases the odds of failing some "less interesting" classes down the line.

Another problem that I've seen with a lot of really clever folks is that if you're told your entire childhood you're smarter than others, but you see these "others" sometimes get more successful, it's really easy to fall into profound cynicism. You never try anything and just undermine others on the internet.

Ultimately, stories like this are an exception, not a rule, even for kids who are truly brilliant. And yeah, it's easy to underestimate the role the parents play, mostly in creating the right opportunities and instilling the right way of thinking about the world. A child doesn't learn to play piano at the age of eight unless there's a piano in the home and a family member or a paid tutor to show them the ropes. Even for stuff like math, it's a parent's choice to buy the right books versus just giving the kid a smartphone.

Funnily enough, I think it was also one of the early warnings of what's coming. The problem with the Flickr algorithm was that it ended up promoting a very predictable style of photos - mostly HDR landscapes - which in turn encouraged more photographers to emulate it. In its early years, it was a fantastic place to learn photography. In the "final" years, it was extremely homogenous and, frankly, boring.

Is this your experience? Do you think 30% of your friends or family members can't answer this question? If not, do you think your friends or family are all better than the general population?

I'd look for explanations elsewhere. This was an online survey done by a company that doesn't specialize in surveys. The results likely include plenty of people who were just messing around, cases of simple miscommunication (e.g., asking a person who doesn't speak English well), misclicks, or not even reaching a human in the first place (no shortage of bots out there).

If you're interested in the user experience, it's this: https://www.reddit.com/r/MySingingMonsters/comments/1dxug04/... - apparently, some annoying ad-like interstitial that many people probably just click through at random.

> spending 5 hours writing a tool to automate 5 minutes of work

Hey, that's a legitimate engineering activity.

> Is it actually being sold at a steep discount? Anthropic CEO has stated they have high margins on inference, so training is the big cost center.

They're spending more than they're making. For the foreseeable future, saying "we could be profitable if we stopped training" if goofy, because they can't stop. If they do, no one will want to use their product because it will be overtaken by competitors within three months.

I get it that in 10 years all of this might peak and we're gonna be content using old models, but that'll be a very different landscape and Anthropic might not be a part of it anymore if they don't start making money before that.

Most hardware drivers are simpler than people expect. The hardware is usually designed to do the sensible thing in a straightforward way, and you're just translating what the OS wants into a bunch of bits you need to write to the right hardware register.

On the flip side, the perceived barrier is high. Most folks don't have an intuitive sense of how the kernel or "bare metal" environment differs from userland. How do you allocate memory? Can you just printf() a debug message? How to debug if it freezes or crashes? All of these questions have pretty straightforward answers, but it means you need to set aside time to learn.

So, I wouldn't downplay the value of AI for the same reason I wouldn't downplay it with normal coding. It doesn't need to do anything clever to be useful.

That said, for the same reasonss, it's harder to set up a good agent loop here, and the quality standard you're aiming for must be much higher than with a web app, because the failure mode isn't a JavaScript error, but possibly a hard hang.

OP, I don't want to be a jerk, but I can't help but wonder why for every other AI promoter, the workflow here extends to using AI to write articles about AI?...

I don't get it. If I wanted to know what ChatGPT / Claude / Gemini think about this topic, I would have asked them.

Nah. Wikipedia is popular because it is the #1 search result for a lot of stuff. Most of people going there just want to look up something for a homework assignment, online argument, or whatever. If Grokipedia has an error rate 5%, compared to 1% for Wikipedia, it's probably still fine.

If Wikipedia traffic shrinks down just to the true "encyclopedia reader" crowd, they will be in trouble, because I suspect that's less than 10% of their current donations. And Grokipedia is already starting to crop up in search results.

Stick-frame single-family construction is pretty inexpensive. For a single-story structure, rough framing and roof can be done by a couple of low-wage laborers with simple tools in a week.

Prefabrication doesn't save all that much money: you still need several guys, now with a crane, and you need to ship and deliver an oversize load. A lot of the cost is finish (including cabinets, countertops, appliances), and in California, land and compliance.

Right, but the reason that Conservapedia fizzled out is that you can't really build a critical mass of human editors if the only reason your site exists is that you have a very specific view on dinosaurs and homosexuality (even among hardline conservatives, most will not share your views).

What's different with Grokipedia is that you now have an army of robots who can put a Young Earth spin on a million articles overnight.

I do think that as it is, Grokipedia is a threat to Wikipedia because the complaints about accuracy don't matter to most people. And if you're in the not-too-unpopular camp that the cure to the subtle left-wing bias of Wikipedia is robotically injecting more egregious right-wing bias, the project is up your alley.

The best hope for the survival of Wikipedia is that everyone else gets the same idea and we end up with 50 politically-motivated forks at each others' throats, with Wikipedia being the comparatively normal, mainstream choice.

"A real number in which each decimal digit at position n is equal to the first digit of BB(n)."

Since you asserted that individual BB(n) numbers are computable, I think you will have no difficulty writing an algorithm that outputs that.

I see some folks here defending Altman because it was an off-the-cuff remark in front of a receptive audience. But why does this make the comment acceptable? Would you give me if I talked about eating babies, but defended myself by saying that I was speaking to a receptive audience?

Most charitably, it's a dumb thing to say. It compares two unrelated things if you see the value of human life to be more than just answering prompts. Less charitably, the argument is evil: if he was trying to make a sincere apples-to-apples comparison, it implies that he doesn't value human life beyond the labor his company can automate.

I can understand edgy teenagers making arguments like that on LessWrong forums, but Altman ought to know better. He either doesn't, or he sincerely believes what the comment implies.

> Would armed resistance be more effective?

I mean, with dictators, that's usually what it comes down to. But it often takes years or decades of unrest and repression before someone with enough guns decides they want to be on the right side of history.

It's a fascinating if morbid process we go through every now and then... sort of, building consensus by sacrificing livelihoods and lives.

Iran is one of the most oppressive regimes remaining on this planet, so I really hope this does it. The problem is that revolutionary governments are usually not dumb and do their best to make sure that another revolution can't overthrow them too easily - hardline loyalists with benefits in the military, etc. So this probably ends with a military intervention by other countries or some other sequence of events that will spell even more misery.

The whole history of the Iranian revolution is pretty wacky. It's easy to take a knee-jerk position that "the West did it", and we definitely set some pieces in motion, but Iran wasn't really hurting prior to the revolution, which is why it caught everyone by surprise. The shah made a number of political missteps, there was some sentiment against the UK and the US, and people wanted change... but almost no one wanted a theocratic dictatorship instead. And yet...

Busy beavers are a classic example. They're mostly-hypothetical numbers that tell you "if any Turing machine of size s runs for longer than this, it doesn't halt." There's a link to that in the sentence you quoted.

Why are tech specs relevant here? The problem with Flock is that once the data is collected, and once it's made accessible to law enforcement without any legal review, it's going to be used for solving heinous crimes, for keeping tabs on a vocal critic of the police commissioner, and for checking what the officer's ex-wife is up to.

If the cameras were installed and operated by the DHS or by the local PD, would that make you feel better? The data should not exist, or if it must, it shouldn't be accessible without court approval. The model you're proposing doesn't ensure that; in fact, it moves it closer to the parties most likely to misuse it.

Vulnerability research is exciting and profitable, but it has three problems. First, it's mentally exhausting. Second, the income it generates is very unpredictable. Third, it's sort of... futile. You can find 1,000 vulnerabilities and nothing changes.

So yeah, it's the domain of young folks, often from countries where $10k or $100k goes much farther than in the US. But what happens to vulnerability researchers once they turn 35? They often end up building product security programs or products to move the needle, often out of the limelight. They're the ones who write checks to the young uns to test these defenses and find more bugs, and they're the ones who will be making the call to augment internal or external testing with LLMs.

And FWIW, the fact that the NSA or the SVR now need to pay millions for a good weaponized zero day is a testament to this "boomer" work being quite meaningful.

> There's an undertone of self-soothing "AI will leverage me, not replace me",

Which is especially hilarious given that this article is largely or entirely LLM-generated.

I can sympathize with this take, but I think it misses the point: the platform is not broken. It's delivering people precisely what they want. If you look at the version of this for young people - TikTok, Snapchat stories - it's the same thing. Busty models, increasingly AI generated, and various made-up "heartwarming stories" or rage bait. Go to YouTube, and you have more of the same.

This is not even an internet-era thing. Before that, some of the best-selling magazines were basically celebrity gossip. Facebook just found a way to scale it and make more money off of it.

The only thing that surprises me now is that people don't actually mind it if you point out that they're liking, commenting, or resharing AI slop. It doesn't even matter that the story wasn't real. It's enough that the kitten is cute, or whatever.

> The one huge advantage of C is its ubiquity - you can use it on the latest shiny computer / OS / compiler as well as some obscure embedded platform with a compiler that hasn't been updated since 2002.

First, for all platforms supported by mainstream compilers, which includes most of embedded systems (from 8 bit to 64 bit), this is not really a concern. You're cross-compiling on your desktop anyway. You'd need to deliberately install and use gcc from the early 1990s, but no one is forcing you. Are you routinely developing software for systems so niche that they aren't even supported by gcc?

But second, the code you write for the desktop is almost never the code you're gonna run in these environments, so why limit yourself across the board? In most embedded environments, especially legacy ones, you won't even have standard libc.

I really want to support open-source CAD, but it's so hard to take FreeCAD seriously. It reminds me of POV-Ray, which was (and still is) a parametric raytracer. An impressive feat of engineering completely derailed by the choice of a "UI" paradigm that made the simplest things unreasonably hard.

Designing 3D parts is hard enough, and while parametric modeling has uses... come on.

> I still stay with C89 because I know it will be portable anywhere

With respect, that sounds a bit nuts. It's been 37 years since C89; unless you're targeting computers that still have floppy drives, why give up on so many convenience features? Binary prefixes (0b), #embed, defined-width integer types, more flexibility with placing labels, static_assert for compile-time sanity checks, inline functions, declarations wherever you want, complex number support, designated initializers, countless other things that make code easier to write and to read.

Defer falls in roughly the same category. It doesn't add a whole lot of complexity, it's just a small convenience feature that doesn't add any runtime overhead.

> Today's AI can basically compete with the low quality stuff that makes up most of social media, it can't really compete with higher quality stuff

But compete in what sense? It already wins on volume alone, because LLM writing is much cheaper than human writing. If you search for an explanation of a concept in science, engineering, philosophy, or art, the first result is an AI summary, probably followed by five AI-generated pages that crowded out the source material.

If you get your news on HN, a significant proportion of stories that make it to the top are LLM-generated. If you open a newspaper... a lot of them are using LLMs too. LLM-generated books are ubiquitous on Amazon. So what kind of competition / victory are we talking about? The satisfaction of writing better for an audience of none?

I can only assume the grandparent means Google+

This is AI slop. It's LLM-generated text. It says nothing of consequence and doesn't address the thesis in the title at all. It's also illustrated with generated images, complete with smeared / nonsensical text.

I don't understand why an established account submitted it and others are upvoting it.

The LLM that wrote this nearly content-free story doesn't know what it's talking about.

The basic distinction in the infosec industry is that advisories are what you publish to tell customers that you had a bug in your product that might have exposed them or their data to attacks and you want them to take some specific action (e.g., upgrade a package, review logs); while an incident report is what you publish when you know that the damage happened, it involved your infrastructure, and you want to share some details about happened and how you're going to prevent it from happening again.

Because the latter invites a lot more public attention and regulatory scrutiny, a company like Microsoft will go out of their way to stick to advisories whenever possible (or just keep incidents under wraps). It might have happened at some points in their history, but off the top of my head, I don't recall Microsoft ever publishing a first-party security incident report.

> making slides/decks to communicate those thoughts,

That use case is definitely delegated to LLMs by many people. That said, I don't think it translates into linear productivity gains. Most white collar work isn't so fast-paced that if you save an hour making slides, you're going to reap some big productivity benefit. What are you going to do, make five more decks about the same thing? Respond to every email twice? Or just pat yourself on the back and browse Reddit for a while?

It doesn't help that these LLM-generated slides probably contain inaccuracies or other weirdness that someone else will need to fix down the line, so your gains are another person's loss.

> "NEVER use generic AI-generated aesthetics like overused font families (Inter, Roboto, Arial, system fonts), cliched color schemes (particularly purple gradients on white backgrounds), ...

Now, imagine what happens when this prompt becomes popular?

Keep in mind that LLMs are trying to predict the most likely token. If your prompt prohibits the most likely token, they output the next most likely token. So, attempts to force creativity by prohibiting cliches just create another cliche.

Several days ago, someone researched Moltbook and pointed out how similar all the posts are. Something like 10% of them say "my human", etc.

> I am of the opinion that incarceration should be "for the benefit of society"; that the person should be behind bars because because they are a threat to society. If they're done with thing because, they may do it again. And that incarceration should be working helping that person become one that would _not_ repeat that crime. Life sentences should only ever be the case for someone that will always be a threat to society.

Sure, but general deterrence benefits the society, too. Case in point: for many years, California effectively decriminalized petty theft, and it caused a lot of grief to normal people. That's an argument for harsher punishment: even if most shoplifters / porch pirates / smash-and-grab people are not hardened criminals, you want to send a message to anyone contemplating that lifestyle.

To give another example: almost no one gets behind the wheel with the intent to kill. But if you severely punish drunk / negligent driving, more people will pause before doing it.

I am really conflicted about this because yes, I think that an LLM can be an OK writing aid in utilitarian settings. It's probably not going to teach you to write better, but if the goal is just to communicate an idea, an LLM can usually help the average person express it more clearly.

But the critical point is that you need to stay in control. And a lot of people just delegate the entire process to an LLM: "here's a thought I had, write a blog post about it", "write a design doc for a system that does X", "write a book about how AI changed my life". And then they ship it and then outsource the process of making sense of the output and catching errors to others.

It also results in the creation of content that, frankly, shouldn't exist because it has no reason to exist. The number of online content that doesn't say anything at all has absolutely exploded in the past 2-3 years. Including a lot of LLM-generated think pieces about LLMs that grace the hallways of HN.

It's dark outside. I'm sitting in my living room with reasonable indoor lighting. I point my "Pro" phone camera on the tele setting (but no digital zoom) at the wall and take a photo. I zoom in on the capture and there is basically no real pixel detail anywhere in sight. It's all smudged to algorithmically cover up severe sensor noise. Crown moulding edges are not even straight lines and all have weird jaggies.

If I go and grab my full-frame mirrorless, every pixel of the image will be usable at 1:1 crop. "Taking photos indoors and wanting to zoom in and crop" is not an extreme test. The usability of 1:1 crops from cell phones is limited even in daylight.

But yeah, if you look at a photo taken in good light with no cropping and at screen display resolutions, it looks pretty close to the output from a pro mirrorless. My only point is that you get maybe 3-5 good megapixels, not 45+.

And that's before we get into dynamic range, bokeh, etc, all of which phones more or less need to fake to approach the look and feel of photos taken with grown-up cameras.