SnapDragon
3mo ago
I agree, when I worked at Google I remember their security measures being extremely well-thought-out - so much better than the lax approach most tech companies take. However, I DON'T trust their ideological capture. They won't abuse people's information by accident, but I will not be surprised if they start doing it on purpose to their outgroup. And they have the tools to do it en masse.
SnapDragon
11mo ago
Indeed, journalistic standards are loose enough that absolutely anything can be framed to make men look inferior or women victimized.
-
"Men are discriminated against in college admission" -> "Men aren't applying themselves in school"
-
"Women are saved first in emergencies" -> "Men treat women as weak and lacking agency"
-
"Women are admired for their beauty" -> "Women are objectified"
-
"Men commit violence more" -> "Men commit violence more" (no dissonance here!)
-
"Men are more often the victims of violence" -> "Women feel less safe than ever, study finds"
-
"Men die in wars" -> "Women lose their fathers, husbands, sons"
-
"Men commit suicide more" -> "Women attempt suicide more"
-
"Men literally die younger" -> "Women are forced to pay more for health insurance" (honestly, I've admired the twisted brilliance of this framing ever since the Obamacare debates)
SnapDragon
6mo ago
Huh? You posted this as if this article is definitive proof that Israel lies. Was there nothing newer than 55 years old? And all the official data in that article is consistent with a mistake that they immediately acknowledged and apologized for. The rest is a speculative conspiracy theory which, while not impossible, requires both a conjured motive for Israelis to intentionally attack their most important ally and a perfect coverup lasting for two generations.
Are you used to being in some bubble where "everyone knows" that Israel likes to intentionally attack US ships and hospitals, so this link is the kind of "gotcha!" you were hoping for? Or were you just hoping nobody would actually click it?
SnapDragon
7mo ago
Oof. You know you've gone off the far-left deep end when governor Newsom, of all people, is lightly coughing and hinting that this is unaffordable. So now my California tax dollars will be going towards supporting a strike for WGA workers who, in 2020, were earning a bare minimum of $4,546 a week. (I know the numbers in the current contract under negotiation were leaked, but I'm having a hard time finding a good source...? I suspect most of the media is on the side of any union, anywhere, anytime and would very much not like the hoi polloi to find out just how rich these brave freedom fighters actually are.)
SnapDragon
4mo ago
Lockdowns aren't on the pareto frontier of policy options for even diseases significantly deadlier than covid imo, just because rapid development and distribution of technological solutions is possible, but ... covid killed one million people in the united states. Yes, mostly old people, but we're talking about protecting old people here. No reason to pretend otherwise.
Speaking of government policy, I wonder how many lives were lost because we couldn't conduct challenge trials on COVID? It was almost the ideal case - a disease with a rapidly-developed, experimental new vaccine and a large cohort of people (anyone under 40) for which it wasn't threatening. If we were a serious society - genuinely trying to optimize lives saved, rather than performatively closing churches and masking toddlers - I wonder how early we could have rolled out RNA vaccines for the elderly?
SnapDragon
6mo ago
The first thing mentioned in that article is that housing isn't being built because the government is actively getting in its way. Sure, a government deadlock will, sadly, not stop the regulators, but it'll (at least temporarily) stop lawmakers from tossing even more monkey wrenches into an already-completely-dysfunctional system. Also, "new rail systems won't get built" just sounds like the status quo to me...
I mean, I still vividly recall that during the long Obama government shutdown the only way they could actually get us hoi polloi to feel any pain was to actively shut down public parks (requiring more effort than doing nothing). When you're doing a performance review, and the answer to "so what do you do, exactly?" is "as long as you pay me I won't set fire to the building", it's time for that employee to go.
SnapDragon
4mo ago
...are you seriously asking this? I'm not an insect. If you want to claim some observation of insect behavior has even the slightest relevance to human society, the burden of proof's on you.
SnapDragon
7mo ago
There are a lot of really good answers in this thread, reasons why historically unions have been a good idea (even if some notable examples have gone too far), but I want to point out that they almost entirely apply to private-sector unions. In the US we also have truly massive PUBLIC-sector unions, which (as far as I know) there is almost no good justification for. Their power derives from the government, which means that when they "negotiate", the government is the one on both sides of the table (negotiating about money that, as always, isn't theirs). It's always seemed insane to me, but maybe somebody here has a good justification...?
SnapDragon
3mo ago
On most forums, if you're a bad actor waging the culture war, it's probably a decent strategy to post a bunch of links like this that are ridiculous non-sequiturs. Most people are too lazy to follow them and have the (usually reasonable) assumption that what's said in them is being accurately represented. Fortunately, I think The Motte is better than that. Looking forward to guesswho's inevitable (re-)permabanning. We need good leftist posters, but he's not one.
SnapDragon
4mo ago
Hi, bullish ML developer here, who is very familiar with what's going on "under the hood". Maybe try not calling the many, many people who disagree with you idiots? It certainly does not "suck at following all but the simplest of instructions", unless you've raised this subjective metric so high that much of the human race would fail your criterion. And while I agree that the hallucination problem is fundamental to the architecture, it has nothing to do with GPT4's reasoning capabilities or lack thereof. If you actually had a "deep understanding" of what's going on under the hood, you'd be aware of this. It's because GPT4 (the model) and ChatGPT (the intelligent oracle it's trying to predict) are distinct entities which do not match perfectly. GPT4 might reasonably guess that ChatGPT would start a response with "the answer is..." even if GPT4 itself doesn't know the answer ... and then the algorithm picks the next word from GPT4's probability distribution anyway, causing a hallucination. Tuning can help reduce the disparity between these entities, but it seems unlikely that we'll ever get it to work perfectly. A new idea will be needed (like, perhaps, an algorithm that does a directed search on response phrases rather than greedily picking unchangeable words one by one).
To be honest, it sounds like you don't have much experience with ChatGPT4 yourself, and think that the amusing failures you read about on blogs (selected because they are amusing) are representative. Let me try to push back on your selection bias with some fairly typical conversations I've had with it (asking for coding help): 1, 2. These aren't selected to be amusing; ChatGPT4 doesn't get everything right, nor does it fail spectacularly. But it does keep up its end of a detailed, unprecedented conversation with no trouble at all.
SnapDragon
7mo ago
Using race and gender as the overriding factors feels icky to me as well.
Shouldn't it feel icky? It's open racism and sexism, no different than the old days of "XXX need not apply" job postings. Not to mention it would literally be illegal for a private company to hire this way. What's weird to me is that Dem elites are so immersed in identity politics that this doesn't feel icky to any of them.
SnapDragon
10mo ago
Not sure if this has been mentioned before, but on the topic of The Little Mermaid, I am extremely confused by the Rotten Tomatoes score. The "audience score" has been fixed at 95% since launch, which is insanely high. The critics score is a more-believable 67%. Note that the original 1989 cartoon - one of my favorite movies growing up, a gorgeous movie that kickstarted an era of Disney masterpieces - only has an 88% audience score. Also, Peter Pan & Wendy, another woke remake coming out at almost the same time, has an audience score of 11%. And recall that the first time Rotten Tomatoes changed their aggregation algorithm was actually in response to Captain Marvel's "review bombing", another important and controversial Disney movie.
If you click through to the "all audiences" score, it's in the 50% range. And metacritic's audience score is 2.2 out of 10. The justification I've heard in leftist spaces is that the movie's getting review bombed by people who haven't seen it. And there certainly is a wave of hatred for this movie (including from me, because the woke plot changes sound dreadful). How plausible is this? I haven't seen the movie myself, so it's possible that it actually is decent enough for the not-terminally-online normies to enjoy. But even using that explanation, how is 95% possible?
Right now I only see two possibilities:
-
Rotten Tomatoes has stopped caring about their long-term credibility, and they're happy to put their finger on the scale in a RIDICULOUSLY obvious way for movies that are important to the Hollywood machine. I should stop trusting them completely and go to Metacritic.
-
People like me who have become super sensitive to wokeness already knew they'd hate the movie and didn't see it; for the "verified" audience, TLM is actually VERY enjoyable, and the 95% rating is real.
But, to be honest, I would have put a low prior on BOTH of these possibilities before TLM came out. Is there a third that I'm missing?
SnapDragon
4mo ago
I mostly agree with you, but I want to push back on your hyperbole.
First, I don't think doing RLHF on an LLM is anything like torture (an LLM doesn't have any kind of conscious mind, let alone the ability to feel pain, frustration, or boredom). I think you're probably not being serious when you say that, but the problem is there's a legitimate risk that at some point we WILL start committing AI atrocities (inflicting suffering on a model for a subjective eternity) without even knowing it. There may even be some people/companies who end up committing atrocities intentionally, because not everyone agrees that digital sentience has moral worth. Let's not muddy the waters by calling a thing we dislike (i.e. censorship) "torture".
Second, we should not wish a "I have no mouth and I must scream" outcome on anybody - and I really do mean anybody. Hitler himself doesn't come close to deserving a fate like that. It's (literally) unimaginable how much suffering someone could be subjected to in a sufficiently advanced technological future. It doesn't require Roko's Basilisk or even a rogue AI. What societal protections will we have in place to protect people if/when technology gets to the point where minds can be manipulated like code?
Sigh. And part of the problem is that this all sounds too much like sci-fi for anyone to take it seriously right now. Even I feel a little silly saying it. I just hope it keeps sounding silly throughout my lifetime.
SnapDragon
3mo ago
"Brutally" slaughtering a pig in "disgusting" "industrial" conditions? Those are very subjective words. The pig doesn't care that it's not being given a dignified sendoff by its loving family at the end of a fulfilled life in a beautiful grassy glade with dandelions wafting in the breeze. Humans fear death; animals don't even understand the concept. As long as we kill them quickly, I really don't give a shit how it's done.
Which isn't to say I don't have concerns about factory farming. The rest of the pig's life may be filled with suffering, and (IMO) we're rich enough, as a society, to do better. My morality-o-meter is ok with sacrificing, say, 0.01% of value to humans to improve the life of pigs by 500%.
SnapDragon
6mo ago
Adding to the list, there's Robert Ethan Saylor, who had Down's syndrome and suffocated after being forcibly restrained by authorities. His crime was slipping back into a theatre to watch the same movie twice. A pretty similar situation to George Floyd, except one was a career criminal on meth, and one was mentally disabled. But we know which one got the national outrage. (To be clear, both just seem like unfortunate, preventable-in-hindsight accidents to me. It's just the hypocrisy that I hate.)
SnapDragon
7mo ago
I'm a Putnam winner, and I don't think it's all that rarefied a category. I certainly don't dismiss out of hand the idea that Elon might be smarter than me. I'm probably better than him at math/programming, but I devoted my life to it and Elon didn't. If he'd had a different set of obsessions, maybe he'd have topped some other category instead of "richest man on Earth". (Heck, I wonder how many pro gaming champions might have been Elon - or a Fields Medalist - with a slightly different set of priorities...)
SnapDragon
3mo ago
I'm far from an expert (and I doubt anyone else in this thread is either), but I'm not sure I really agree with your "extremely dangerous" assessment. Lots of things have a 100% kill rate. Like, congratulations, they've reinvented rabies? A virus that represents a serious risk to society needs to combine a number of unlikely factors, and "killing the host" is probably the easy part. (Ironically, after a certain point, high lethality makes a virus less threatening - a virus's host needs to survive to spread it on!) To truly threaten civilization, you'd have to combine it with a long asymptomatic but highly contagious incubation period.
Of course, because the media are idiots, the article you linked mentions the "surprisingly rapid" death of the mice as if that's supposed to make it more, not less, scary. Ah, journalists, never change.
SnapDragon
4mo ago
How so? He could have just retired slightly sooner, still quite rich and still doing the rounds on news/talk shows answering tough questions like "how does it feel to have saved eleventy-trillion lives with Science(tm)?" Rand Paul constantly grilling him would barely even be reported on, let alone actually affect his life.
SnapDragon
3mo ago
Citation needed...? It's a little hard to ask the pig. And even if true, should I care overmuch that the pig "feels stressed" for the last hour of its life? Humans go through worse (to say nothing of how animals die in nature!). If you want me to care about animal welfare, you should focus on the part that really matters - the life the pig lived - rather than the lurid, but ultimately unimportant, details of its death.
SnapDragon
7mo ago
Huh? The primary selection criterion, stated clearly and up front by Newsom, was "is a black woman". All other considerations, including the unobjectionable non-icky one you just changed the subject to, were secondary.
SnapDragon
3mo ago
Maybe I'm missing some brilliant research out there, but my impression is we scientifically understand what "pain" actually is about as well as we understand what "consciousness" actually is. If you run a client app and it tries and fails to contact a server, is that "pain"? If you give an LLM some text that makes very little sense so it outputs gibberish, is it feeling "pain"? Seems like you could potentially draw out a spectrum of frustrated complex systems that includes silly examples like those all the way up to mosquitos, shrimp, octopuses, cattle, pigs, and humans.
It'd be nice if we could figure out a reasonable compromise for how "complex" a brain needs to be before its pain matters. It really seems like shrimp or insects should fall below that line. But it's like abortion limits - you should pick SOME value in the middle somewhere (it's ridiculous to go all the way to the extremes), but that doesn't mean it's the only correct moral choice.
SnapDragon
5mo ago
Not only do you get to use "think of the children", you also get to partake in socially-approved hate for a group of weirdos for their innate characteristics. Humans have always had an appetite for doing this, but in modern times there are far fewer acceptable targets.
SnapDragon
10mo ago
...she didn't intend to expose confidential or classified information and most of the email saga came down to a mixture of negligence and pride...
...It was the pride of "owning" them...
Um, your justification seems to apply equally to both Hillary and Trump, even by your own words. So far as I know, nobody is accusing Trump of actually intending to expose the information.
SnapDragon
10mo ago
I'd be ambivalent if it was just a few instances, but it really feels like he's exploiting the system. I wouldn't come to themotte if every other top-level post was one person soapboxing about da joos. HBD was similar: Yes, this is (intended to be) one of the few places on the Internet you can freely debate it, but it shouldn't be the only topic of discussion...
And that's the point. That's the one, last, important step that (much of) science is lacking. Have you built something that works AT ALL? It's not that engineering doesn't suck. It's that modern "science" is even worse, because so much of its product (random unreplicated research papers, written on esoteric subjects, skimmed by friendly peer reviewers and read by nobody else) never needs to pass that final filter.
More options
Context Copy link