I am absolutely baffled at the idea that LLMs mean you need less automated verification of correctness.
3,730 karma · joined November 6, 2018
The ability for a handful of companies, simultaneously very powerful and easily susceptible to pressure from other powerful actors, to do the same sort of thing with the next generation of core learning and engineering tools, is freaking terrifying.
It's entirely possible an EU country did this; they're only vaguely guessing Belarus or whoever. In most countries, it's a big deal if the spies are caught spying on the domestic government.
> quite a few private entities do as well.
It's a risky game, doing that. You don't get any of the professional courtesies, and you're not usually eligible for the prisoner exchanges.
That's because you lack imagination. 99 percent chance they are blocking you because they don't want "divisive political rhetoric" on the platform. Allowing a surveillance state is "apolitical" as long as it doesn't involve rocking any boats or making any noise.
... and NSLs don't do that. It would really be nice if people actually understood what NSLs were before blaming everything on them. Trust me, they are bad enough without inventing stuff.
... and information isn't really the question. Not that there's actually any good definition for "social media".
It is far more common for that child to be targeted by parents, and maybe by people they know in person, especially because of the lousy social environment their parents have pushed them into, and therefore to have limited offline support systems, and you are now trying to take away all they do have.
Maybe if it's some server-side software that you only use yourself...
If you want to be a piece of critical infrastructure, you need to deal with the implications of that. It's not OK for private entities to be able to "unperson" people in important parts of their lives for what amounts to convenience reasons. If them not being able to do that raises the price of the service, so be it.
Honestly even banks, which are already highly regulated and at least have more nominal competition, still have too much leeway to cut off customers based on error-prone statistical methods, without recourse or explanation.
In fact, I don't actually use it, but as an experiment I once set up and fooled around with an OpenRouter account over Tor. It did demand an email address, and I gave it a Proton account also set up over Tor. Both were paid for with anonymous cryptocurrency: Monero gatewayed via some random exchanger.
Whereas I never signed up for an Anthropic account because the first screen I hit demanded a phone number. I mean that was the only thing on the screen, and you weren't going anywhere until you provided it. It's been quite a while, though.
Perhaps there are different paths to getting accounts.
Montreal. I think there've been three thunderstorms in the past week, none of which I would have predicted if I'd stepped outside a few hours beforehand. Yes, that's an unusual week, but I've been caught in storms often enough, and you can't build your routine around conditions you can't rely on.
And only absolute fanatics ride bicycles in the winter. Even though I think the facilities are pretty good.
> From what I've observed, the vast majority of car trips in my city could be easily replaced with a bicycle trip.
You observe people driving around in cars. You have no way to know where they're going, where else they've been or may be going on the next leg of the trip, what they're carrying, anything about their physical condition, or, well, anything other than that they're driving. Even if you did a formal "study", you wouldn't know if you were asking the right questions (not that the people doing "studies" usually even try).
Personally, if I go anywhere at all, I'm usually moving more cargo than I could move on a bicycle, and doing that is the whole purpose of the trip. Cargo or no, it's often to a place I could not practically reach on a bicycle. And when neither is true, I can usually walk. The number of trips where I could use a bicycle is not large enough to make it worth maintaining one, and that wouldn't change no matter how good the infrastructure got.
Also, for a good chunk of the year, it's icy enough here that basically nobody rides bicycles... and in the rest of the year, it's not uncommon to get unpredictable thunderstorms. Sorry, riding in that is not "enjoying life".
By the way, the economy that feeds you and maintains your bicycle also depends on motorized road vehicles and the infrastructure to support them. The converse is not true; you and your bike could go away and nobody else would notice.
Seriously, they're very, very limited, niche vehicles. If your life fits into that narrow niche, then good for you... but the rest of us are sick of hearing from you.
The reason to regulate in maybe 2000 or so was that staying with IPv4 led to NAT. NAT led to it being impossible for users to receive incoming connections. Inability to receive incoming connections led to (a) horrendous protocol complexity, (b) probably some applications never even being invented, and, (c) everybody using ultra-centralized services. Ultra-centralized services led to advertising-driven distortions of service utility, concentration of political and economic power, and choke points. Choke points led to surveillance state bullshit that's just fully ripening today.
And, yes, this was (in broad outline) foreseeable in 2000. I wasn't the only one.
... but that only works if you design properly from day one.
How about no? They'll block traffic there, too.
Sure, whatever. Maybe in some ways.
> I personally know children who've posted too much, and children who've been solicited directly by adults, both to try and meet and for nudes.
... but not in that way.
I personally knew children who'd been solicited directly by adults before there was even an Internet. Including me, if you use the definition of "child" that seems to be popular in this sort of debate (and, by the way, it wasn't a big deal).
We did not shut down the world because of it.
Yeah. Bluesky.
Pretty clear you have no clue and shouldn't be listened to.
Random anecdotal claims from population A contradict random anecdotal claims from population B.
> I’m curious what the author has in terms of actual evidence here?
Well, it can't be any worse than you have, in that the paper you link to doesn't show anything about what causes that negative Flynn effect. It does speculate, and social media is not on even on the authors' list of guesses.
Did you have anything relevant?
While increasing harm to many others. Nobody ever wants to mention that part. It is not a clear "child safety" win. Personally I think it's probably a significant net loss.
There are of course also people who are in it for the universal surveillance and see the others as useful idiots, but the idiots are genuine.
We're not assuming it. We're observing what people are in fact actually deploying.
Yes.
On at least Android phones, and I'm pretty sure also on iOS, location access is a user-controlled permission that's not necessarily granted to any given app. There always leaks, but are you going to require commercial software to play cat-and-mouse games to get around system security settings?
And some, possibly most, "social media" can also be used without apps.