I really hate the way the word “nihilist” is used in modern English.

That aside - my point is that this was expected.

And they didn’t live during 30 years war or whatever they remembered as scary. Didn’t matter.

WDYM turned out? Reading about it, seems to just be a long-running project. Still in construction, nothing to see there, such stuff.

Why not, WWI and WWII and things around them outdid all the scary stuff people were predicting just fine.

WDYM why not drugs? Strong alcohol and tobacco were a state monopoly in many countries for many years, and still are that kind of business heavily intertwined with state.

And prohibited drugs are in a similar position, just the state monopoly is unofficial. Well, officially they are prohibited. As I said. OK.

There are plenty of different kinds of drugs with different degrees of addiction, lethality, damage to the environment in production too.

Anyway, about AI - were it not profitable, it wouldn’t be rolling. It might be a bubble, but it’s, for the users of this kind of technology, a qualitative change from precise algorithms and industrial optimization to fuzzy decisions based on statistics. Basically machines following human orders as humans mean them. Well, kinda like that, except very computationally expensive and as good as your dataset, but there are tasks hard to do the old way and easy to do the new way. Hence (on Jingle Bells melody) bombing swarms, bombing swarms, bombing all the way. I suppose not all the way, but the essence is here. Also surveillance, detecting and stopping people harmful for some political end without ever alarming them or anyone or using direct visible force, predicting events.

It’s inefficient when used for the same tasks a shell script can do. It starts being efficient when used on the scale of lives and families and groups and communities, and armies and economies. Matter of scale.

In that light I’m glad I never played that, living in Moscow near an internal troops division. Doesn’t matter, though, someone has.

Plus, it’s disgusting and should be illegal anywhere that it isn’t just in general. It’s weird that you’re defending it like it’s diet coke or something.

“It’s disgusting” is not quite the right argument for making something illegal.

And that “you’re defending” presupposition should honestly be your last claim in any group of people before being shown the door.

You seem to have that “all or nothing” mindset in an argument, as if you really didn’t like someone, then they should be prosecuted as a rapist, a murderer and an arsonist at the same time. Exaggerating, of course.

Quoting myself.

I “honestly think” each case is unique. Just like with everything else.

CP is harmful due to normalizing the thing, useful due to redirecting some of the energy people with that pathology have away from, you know, real children.

Well. To hang pictures and hooks it makes sense to have wall panels with holes or whatever. And not drilling bearing walls.

Small-small price might be even fine, it’s scary how we live in a world full of wireless connectivity. Just some wire between rooms, LOL.

Just - if attaching stuff with magnets isn’t generally common (except fridge doors), then making magnetic concrete walls doesn’t seem to have potential in helping with that.

When I saw that headline, I was thinking of something like concrete magnetic floors in space, something sci-fi.

Flying concrete from Earth’s gravity well seems very expensive, though.

it’s essentially iterating on known images

No. It’s iterating on the common traits of known images compressed plus lots of randomization.

If you trained a Stable Diffusion model on only pictures of Rwandan people, and asked for an image of “a man sitting on a chair” the man will look vaguely Rwandan.

If you train a model on adult pornography and non-pornography with children and adults alike, it might be capable of generating plausible child pornography.

When you train an AI on CSAM, it produces images that are based on CSAM. Real people were victims in the base material, too. Close e-fuckin’-nough. Real people’s victimization is literally the core of how those images are made.

I’ve just told you how this is not true.

You seem to have that “all or nothing” mindset in an argument, as if you really didn’t like someone, then they should be prosecuted as a rapist, a murderer and an arsonist at the same time. Exaggerating, of course.

Point being that child pornography without real victims is something not contested here and has its own implications. You are trying to argue on something out of reach.

And he started with police officers being supervised, but the real problem is not what we all see about an event, it’s whether we who see it one way have power over those who pretend they saw something else. So if those deciding also don’t want to reduce police officers’ loyalty to their superiors and readiness to obey on the job, they might find an excuse for a clear murder in FHD.

Usually the place of origin of a technology is also where most its uses are pioneered.

as much as Grok is a CSAM machine if you pay for it.

CSAM is Child Sexual Assault Media, and Grok is not providing that, it’s providing Child Pornography.

You are comparing making non-consensual material with real people to generating material with no real people (based off real media, though, but that’s an implication with everything AI-generated).

Non-malicious compliance would be a protocol extension, don’t ask me how, but if WebSockets exist, then it’s possible to make an EuHTTP standard to which you’d upgrade. So that all these popups wouldn’t be needed and you’d conveniently set things up on the client.

Actually owch. One can just take some WS library and make a Gemini-like protocol, only over WebSockets (allowing for much of normal infrastructure to support it, you know, nginx, haproxy, lots of stuff), that would leverage convenient existing technologies and without need for Google’s browser engine more complex and expensive than a rocket.

OK, that’s called NOSTR, they are just not aiming for replacing Web in any form. For now.

EDIT: And this probably is not what’s being discussed.

The whole global auto industry should be incentivized to go back to that “runs forever” design focus of the Mercedes W123 series and improve on it with more longevity, cheaper serviceability.

That works when the car itself is something produced in small batches and a very capital design and costs accordingly, or when its maintenance makes bank and for the producer at that, or when there’s continued growth, so you don’t sell new cars to people with old cars.

I mean, of course there’s the variant of cars being as modular as PCs and a Mercedes of Theseus being possible. Always profitable for the producer, since from time to time the customer buys spare parts (a law is necessary that it’s legal to make and sell spare parts for anyone ; just like with Apple stuff, official things will be more popular), and never just fully going to junk at once. Seems the most realistic variant for me, economically, of the good ones.

There’s another variant, a dystopian one, being implemented in fact, where car producers own you via parts pairing, planned obsolescence, parts barely surviving guarantee age and impeded repair and telemetry, all at once.

Without ability to put pressure almost like in war, the modular variant would be the equilibrium, unfortunately we the humanity haven’t yet adjusted our societies for computers (no need for a more complex description).

There are different kinds of “old cars”, the kind of old cars made before 70s that are really inefficient with gasoline, but might last another hundred years if maintained, and the kind of old cars made up to 90s that are harder to keep from falling apart, and then the kind made later, which is - not really for future generations.

The more optimized their production is and the less luxurious they are as a thing, the closer they are to something that’ll only last their guaranteed time. Preferably for the producer - falling apart into rust a couple of days after that.

which is probably better than diverging hallucinations

Shouldn’t it be the other way around?

Telegram would be just wonderful were it being marketed as what it is. It’s a gorgeous mass groupchat system.

Nothing private at all, but for that you get convenience.

And I would like something also private and still fit for mass groupchats, I don’t know, perhaps, instead of encrypting messages for every participant have some kind of rotating symmetric keys for everyone, like with encrypted TV streams, signed by a smaller set of group moderators. That could fulfill the same role and also be peer-to-peer.

But a lot of things exist beyond our imagination, it’s just that for something to be persistent someone needs to make money on it.

… lost in recent like 20 years, in my childhood dissent was far more normal than now. I think it’s social media, but also the commonly pursued concept of what’s socially respectable, solid and real.

And dissent is, by normalization, what allows peace in society.

I’ve read this happened because sometime in the 80s comrade Reagan decided to own the Japanese instead of letting competition do its work (for cars, but with electronics similar things followed). He’s somehow often associated with liberal capitalism and so on, but the guy believed that “monopolies are efficient”, but at the same time by some magic if a monopoly stops being efficient, then all the capital and technology base for competition to replace it will just materialize in one place in one moment all by themselves. So I’m not even sure if “comrade” is irony. The ironic part is that the US president whose term coincided with Soviet system conclusively losing the Cold War is also the one who supported state capitalism and ideologic pressure in society.

And then we no longer have responsible humans with their set of human connections, which are a burden in the sense of corruption and bias, but also the reason the system kinda works, because a human with their friends and relatives at least care to try to avoid jail or infamy.

That wouldn’t work well. You either have that role as a node in the graph or as a responsible agent. In the former case its functionality should be very precisely defined and have no social contention value, which is not true for courts and judges, they are actually half of that social contention. In the latter case responsibility should be clear, that is, a computer program acting as that clerk or higher should transfer same and full legal responsibility to everyone in chain operating it, which is not the case now.

At the same time - nothing really new, power makes rules, the only important part is how fine-grained friction is, which leads to some equilibrium. More fine-grained is gentler, but also benefits those with most power. More coarse-grained means conflict, and eventually still benefits those with most power, but approaches the equilibrium faster. I suppose at this point we’d want the latter, and then the former.

Because, well, these changes erode social systems that had been building up for centuries, so faster is more important. I think.