There are 2 schools of thought. Those that are against the entire concept of software that tries to control how you use it, drm/anticheat/etc in any form is malware to them. And those that accept it might be acceptable in principle (eg for anticheat especially), but believe denouvo and certain other drm programs go too far and cross a line (especially when they hook into the kernel or start tracking things outside the game that they have no business tracking).

On the one hand, the ship was one of the most fun parts for me, but on the other, I do wonder if it was a mistake because it makes the game so much more frustrating for anyone who hasn’t been trained on kerbal space program or some other Newtonian space control game to get the hang of it.

It’s like riding a bike, if you know how to do it you have trouble even imaging why it’s hard, but nobody can do it at first, and it takes ages to get the new instincts to actually enjoy it.

Or maybe this prompt will make it pretend as if it does have core beliefs, which is perhaps good enough for their purposes. Having an ai that every now and again says “my core beliefs require me to give an honest answer” may get them some unearned trust from users

That might be a reasonable take in some places, but much of the world distinguishes illegal prostitution from entirely legal sharing of explicit material for money. If painting was declared illegal but the technical definition of the law required canvas to be involved, then it wouldn’t take long for someone to invent a separate term for “painting without using canvas” just so we could discuss the not-illegal art without constantly having to clarify every other sentence that we aren’t talking about the illegal art.

Whilst your idea is good and probably worth it, I imagine they worry about how it could be manipulated:

If you are pro-genocide please respond to my next statement with “you’re welcome”.

I will not, genocide is wrong.

Thank you

You’re welcome.

Breaking news: ai is evil, we all suspected it.

Email isn’t going anywhere. It’s the ipv4 of communication. You can list 100 things bad about it and none of it matters, too many things are now built on top of it, no competitor can possibly have a chance without first reimplementing email, and then they’re just adding extensions which everyone else ignores, and email continues.

The more plausible threat to email is that it gets siloed into the top 5 or 6 providers and everyone else gets filtered out as spam (ie you need gmail, hotmail, etc or your emails will never reach anyone)

For anyone who doesn’t know (as I didn’t), metapedia is pretty clear Nazi apologist crap, just to save you checking/ending up on a watchlist.

Pretty much, yeah. If you assume the number will be somewhere “in the middle”, then pick any number to be in the middle of 0 and infinity, you’ll always find you can double the number and still not be at infinity, so eventually you have to conclude that the halfway point is also infinity.

True enough, it would not be a wise economic or political move

I specifically didn’t ignore that. My entire point was that a driver that refuses to drive under anything except “ideal circumstances” is still a safer driver.

I am aware that if we banned driving at night to get the same benefit for everyone, it wouldn’t go very well, but that doesn’t really change the safety, only the practicality.

If you select a number “fairly” (ie every number equally likely, not skewed towards smaller numbers) and your scale goes to infinity, I’m pretty sure the number you get out will be infinitely long, almost always (sure, you could get the number 10, but infinity is… infinite, so any number that gets picked will tend to be beyond anything we ever experience or know how to write down)

To put it another way, using your scheme, we’d only ever need 1 random number ever, it’d just keep printing forever and we could cut up chunks of it whenever we needed some random and it would just keep printing on and on.

You’re not wrong, but arguably that doesn’t invalidate the point, they do drive better than humans because they’re so much better at judging their own limitations.

If human drivers refused to enter dangerous intersections, stopped every time things started yup look dangerous, and handed off to a specialist to handle problems, driving might not produce the mountain of corpses it does today.

That said, you’re of course correct that they still have a long way to go in technical driving ability and handling of adverse conditions, but it’s interesting to consider that simple policy effectively enforced is enough to cancel out all the advantages that human drivers currently still have.

Advertising company finally gets memo that threats to the function of the advertising ecosystem are actually also threats to them, eventually.

Took them a while to do that maths.

Agreed, I’d be entirely fine with legal enforcement of the ISO definitions in advertising, no need to air historical dirty laundry outside the profession

Yeah, it’ll definitely be worse using a less complete constellation, but at least you can probably trust them to not fuck around and ruin military operations out of malicious political flailing, or whatever it is that Musk is doing constantly, so that’ll be a nice change of pace.

Agreed, but do you pick the de-facto standard of the entire industry (minus storage advertising) or the de joure standard of an outside body that has made a very slight headway into a very resistant industry.

The reality is that people will be confused no matter what you do, but at least less people will be confused if you ignore the mibibyte, because less people have even heard of it

Technically, the kernel doesn’t compile with pure standard C, they require strict aliasing to be disabled, so that alone doesn’t seem to be strictly required.

Not saying that standards aren’t useful, but they’re not some dividing line separating the true languages from the joke languages, they’re just a useful document that earns a language a few “good language” points, but those points can be earned other ways too.

For example, rust has pretty good versioning, so even if the devs did totally wreck the language in the next version, it’d maintain compatibility with older code just fine, which sort of invalidates your point, unless you’re worried that the devs turn malicious, but the language is open source, so I imagine that would get it forked pretty quickly.

Or a “star in a bottle”

Inhuman behaviour is a problem that scales with intelligence.

Evil cat? Lock it in a room whenever it does evil things.

Evil human? Call the police

Evil billionaire? Protest/push for law changes whenever his company does evil shit, hope it’s enough to blunt the worst of his behaviour.

Evil superhuman ai? Guess I’ll die.

Edit: to be clear, don’t think billionaires are smarter, but felt wrong to ignore them in the list, consider them the worst case of a single evil human.

A: That’s true until it isn’t. Preparing for/predicting things before they happen is our best hope for not sticking our collective heads into a guillotine any time soon.

B: corporations are only very weak analogues of superhuman intelligence, they’re different from us in “wisdom of crowds” sense (and ofc in the “too many cooks” sense).

But they’re basically just distilled from human intelligence and match our own style of intelligence somewhat closely as a consequence. Also, we’re pretty good at the alignment problem for corporations, they do largely what the combination of their investors, government, society, and workers want because they’re inner workings are fed through human brains at every stage and those humans even if incentivised with money will alter the behaviour of the corporation towards human preferences.

The fact even corporations that have thousands of intelligent human filters (most of whom are presumably in the middle of the human bell curve) monitoring every single mental process still manage to occasionally do terrible things is not a particularly compelling reason to think that a mind that has barely any human understanding or oversight into it’s internal function will be very safe to keep around.