lacewing

> They claim the AI itself built an OS and demoed Doom running on it.

Doom ran on MS-DOS, which - by modern standards - provided a shockingly minimal set of abstractions for programs. I think about the only thing you need to run Doom is the int 21h "API" to access the FAT filesystem and perform keyboard I/O. Note that MS-DOS did not provide facilities such as memory virtualization / management, process management, video drivers, sound drivers, etc - that was all provided by the hardware itself, which had its own hardware interrupts handled by the code in the device's ROM. It's why Doom required you to choose the type of a sound card you have, the interrupt / DMA channel to use, etc.

So I think this is a lot less of a flex than it seems; in fact, Anthropic using agents to build a semi-unusable compiler is far more of an achievement. Providing enough of int 21h to run Doom is probably something that a human could do in a weekend, doubly so if they can peek at the source code of FreeDOS.

In the US, wired phones were often leased from the phone company! If you shop for vintage phones from the 1960s, they will often have a decal to that effect. Examples:

https://i.ebayimg.com/images/g/71gAAeSw1URp8QPF/s-l1600.webp

https://i.ebayimg.com/images/g/tPAAAeSw~RxqBiLR/s-l1600.webp

I think this started to change only in the 1980s and 1990s, thanks to cheaper electronics imported from overseas?

> While you may say it's not television because we have content creators now

What's the difference? There are hundreds of millions of people watching a relatively small number of wealthy creators who need to cater to advertisers and follow strict platform rules or else they get the boot. We even have celebrity gossip and controversies surrounding them.

We've recreated the exact thing we've been fighting.

And if you need to post more than 50 tweets and 200 replies a day, you should probably pay a therapist before you pay Elon. Sincerely, that kind of engagement with total strangers is just terrible in the long haul. I've never seen anyone with that kind of usage who seemed to be happy and well-adjusted.

Your original post said that we shouldn't be worried because people appreciate radio and music for reasons that presumably can't be replicated by AI. I'm asserting that's not true: it's not how most people listen to radio or music, and AI content is already quite prominent.

I'm perfectly familiar with KEXP and other stations like that, but this is not how most people experience the medium. It's like insisting that Taylor Swift will never catch on because her music is not nearly as rich and complex as Wagner. Sure, but that's completely irrelevant.

I think you're talking about some Platonic ideal that just doesn't exist anymore.

Streaming services such as Spotify are increasingly filled with AI-generated songs and the average consumer doesn't seem to mind because we're not listening intently in the first place: it's just a background track we're not really paying attention to. I'm pretty sure that radio execs are looking at that and are taking notes.

For talk radio... if I had a penny every time someone on HN brought up that they're enjoying NotebookLM-generated slopcasts, I'd have a neat pile of coin. And I think it's the same story: most people listen to podcasts just to kill time. Soothing, zero-calorie LLM banter will do.

For a subset of bugs, yes. For some others, not really: I've seen LLMs make bogus assumptions about the threat model (in which case, the exploit works but doesn't demonstrate anything useful) or "cheat" by modifying the code to demonstrate a hallucinated issue.

Frontier models, including Mythos, can greatly streamline bug hunting and exploit developments in the hands of a competent security engineer. In the hands of a person with no security experience, they will still mostly waste your time and money.

I understand that there's more to it than TV ads, but Coca-Cola reportedly spends only about 2x as much, and... you'd agree that they get a lot more bang for the buck in terms of selling the brand?

Yes, I'm clearly on the payroll of... um... big optout?

I'm amazed that Hershey is spending $2B on marketing. I honestly don't think I've seen any advertising for the brand itself at any point in the past decade, and I don't have any positive association with it: it's the "bad" chocolate you buy for cooking?

Looking at their subsidiary brands, I don't think I've seen any serious advertising for them either. KitKat is doing viral marketing on social media, but surely, that's not $2B... I've never seen anything for Twizzlers, Pirate's Booty, Milk Duds, Whoppers... the only reason I know these brands is that they're on the shelf in my grocery store.

Unless I'm an outlier, I think it's smart of them to stop throwing $2B down whatever hole they're throwing it into right now.

No, the system is rigged in a different way. There is an actual data broker industry that's mostly hidden from view; and then there are hundreds upon hundreds of ephemeral "look up your neighbor" websites that you know about and end up submitting opt outs to, only for half of these websites to disappear and a new crop to show up in a year or two.

I can't imagine what possessed the author to write this much text on the topic. But yes, that was always my impression. My pet conspiracy theory - which is completely unsubstantiated - is that if you're pushing for data centers in rural communities, you actually want people to get fixated on water usage, because you can then talk to the county commissioners and shoot down these criticisms so easily.

Mega-scale AI data centers have other externalities. They're often touted as a way for rural counties to become the hubs of the digital era, but they don't employ many people, don't generate a whole lot of tax revenue, and basically just leverage cheap electricity at the expense of local residents. So it's a sham in that respect. You're not gonna have X, Google, Amazon, or Meta reinvigorate your community. You're just gonna have ratepayers subsidize some inference via higher electricity bills.

I have no doubt that someone will chime in saying with an "actually..." that electricity is fungible and therefore, it doesn't matter where the datacenter is built. If it were so, they wouldn't be getting built in places like Wyoming or eastern Washington, and electricity prices in these markets would be the same as they are in the SF Bay Area. In practice, though, there's plenty of factors that make the US electricity markets a lot more local.

Microchip is known for supporting their chips for longer than most. They basically keep making them as long as there's demand. And the Dx series is pretty recent.

That they're flirting with a Cortex-M0 pathway probably suggests they don't want to keep developing another generation of 8-bit platforms after the Dx series. If they bring the same features to another CPU core, I guess that's fine.

I think we're rediscovering management from first principles. The main selling point of AI is that it writes code faster than you could. Checking it line by line undoes most of that benefit. In the same vein, there's no real benefit to leading a team if you plan on supervising every task.

But here's the thing: for humans, this is manageable because we've come up with a number of mechanisms to select for dependable workers and to compel them to behave (carrot and stick: bonuses if you do well, prison if you do something evil). For LLMs, we have none of that. If it deletes your production database, what are you going to do? Have it write an apology letter? I've seen people do that.

So I think that your answer - that you'll lean on your expertise - is not sufficient. If there are no meaningful consequences and no predictability, we probably need to have stronger constraints around input, output, and the actions available to agents.

No. I was answering to parent's broader comment about "formative" writings on nicotine. And I was making my own broader point that the NYT piece is biased, but selective evidence-seeking in the rationalist community doesn't deserve any special praise.

Meh. Writing like this was formative for me (before Gwern; I'm old), but I've come to realize that the biases of the rationalist community are really no different from the biases of anyone else. It just manifests in a different way?

It boils down to an obvious disparity in the standard of proof they demand for "pet" topics versus what they need for everything else. You can do this kind of ultra-nitpicky "rational inquiry" to undermine anything you don't like. You can use it to argue against seatbelts. Or against the ban on lead paint. Was lead paint really all that bad?... and I mean, really? Are there studies? Are they high quality enough?... Double-blind? Confounding factors? Correlation or causation? Even if they look solid, I bet they contain enough errors to cast doubt. Cui bono? What was the role of the titanium dioxide lobby in all this?

For nicotine specifically, I've been around enough people seriously addicted to nicotine to just roll my eyes at this stuff. I had things thrown at me by a visibly jittery relative when I refused to smuggle cigarettes into a hospital. Do I have a published double-blind study showing that it's worse than coffee? No. But again, neither do rationalists for 99% of the stuff they believe in.

Do I think that vapes are a noteworthy problem to be focusing on? Maybe not, but public policy is always to some extent vibe-based. And the harm of being too heavy-handed on vapes is really not something that keeps me up at night.

I used to live in Massachusetts. I'm not sure it's a benchmark to look up to. If buy a brand new car from a dealer, your next stop must be an inspection station - a pointless waste of time and money. But if your car is older than 15 years? You're no longer required to have an annual emissions test. Pretty backwards.

Just for the record, this is a novelty domain with a history of posting AI-generated articles that are clearly designed to get clicks based on nostalgia for how random things "used to be better":

https://news.ycombinator.com/from?site=worseonpurpose.com