Posts
Comments
A basic pledge (that is actually followed) to not delete everyone's data, finetunes, etc.
Such a commitment would not be credible. Even if you're confident Meta won't do the classic Google thing of sunsetting the service, or go broke, or change CEOs, they'd still be subject to things like DMCA takedowns that running locally safeguards you from.
I accuse you of being disingenuous. Zuckerberg's points seems pretty clear to me: from his experience in the closed Apple system, he understands the benefits to developers and end users of having an open system. (Obviously Apple benefits from their closed system; I bet he never feigned confusion about why they impose such restrictions.) ChatGPT and other popular LLMs are closed like the iOS, so he intends Llama to be open like Android/Linux, or if you feel the inscrutable nature of the models precludes that, open like Windows.
The bit about America's advantage being "decentralized and open innovation" of course involves a bit of rah-rah jingoism, but it's a coherent claim: what developers are free to create with open systems like the one he wishes to establish benefits America more than China, even if China would also benefit.
Your "which is it?" suggests you think there is some contradiction to resolve. Developers have a few different options right now, but the reasonable choices are all closed, and if they invest too much time and effort into those, the result of network effects will cause Meta to suffer from essentially vendor lock-in. Hence the need to have an open system that even if not quite as good as the best, is close enough that the flexibility and control that the openness offers makes up for the shortfall.
Zuckerberg's also clearly not the Open Source Zealot you caricature him to be. The Quest is pretty good on openness, but its Horizon OS is not open source, and there are limitations on camera access, and Meta can still close down all sideloading capabilities at will.
Also, to engage in a little bit of mind-reading, Zuckerberg sees no enemies in China, only in OpenAI et al. through the "safety" regulation they can lobby the US government to enact. The plausible catastrophe to work against then is them getting as much of a stranglehold over the AI space as Apple has over its own. Of course, he can't say this out loud without hastening the doom he's trying to avert.
RAAIA is a terrible acronym. How do you even pronounce that? Raya?
The PATRIOT Act of 2001 expired in 2020, so that nifty acronym is up for grabs! If you long for the the same kind of thing, the comfort of knowing the government is watching over you all the time, judging your conduct (like Santa Claus!) and/or the thrill of righteous satisfaction at seeing strangers and enemies who are found wanting be stripped of dangerous freedoms they ought not to have, and for the same kind of reasons, the existential threat posed by Arabs/Iraqis or whatever, reüsing the name wouldn't even be deceptive. Perhaps something like "Promoting Accountability and Targeting Risk In Open-source Technology".
You see the Greeks, but you do not see the horse.
a contract is a contract is a contract, I knew what I agreed to, and I was not about to break my word even if I wasn’t legally bound to it.
This strikes me as a … misguided personal interpretation of legal contracts: you're playing a game straightjacketed by rules of your own devising. Laws are threats, not moral codes.
My personal view is that if I am bound by honor, I will act in good faith, and keep fastidiously to even the spirit of the agreement, even when no one is watching. If I am bound by law (or some other threat), you will get from me only what you can enforce, and only so long I am unwilling to pay the price you threaten to exact.
There was some rule (in my college dorm, I think) everyone was asked nicely to follow, and I did. Then a while later, they introduced a fine for breaking it, and I assessed the risk of that minor fine worth the convenience of breaking the rule, and never followed it again.
Homeless person or professor?
It's a dog-eat-dog world.
This one's better in Latin: Homo homini lupus. Man is a wolf to man.
The people cry out for a strong leader.
This one's better in German.
My favorite adage, from Schiller, is "… the gods themselves contend in vain."
Lest your experience dissuade others from thinking of possibly-great ideas that end up not working out because ideas rarely do, remember the Parable of the Cult of the Rock.
Re: your suggestion to mobilize different groups, I suspect that's still motivated (at least in part) by doing something you can measure. Demobilization, turning opposition voters towards either "protest votes" or abstention, might be cheaper, easier, and more effective.
Alternatively, the "AI safety" movement has made no mistakes of consequence, and has already won decisively.
(Yes, this might basically be the old joke about the Jew reading Der Stürmer, coming from me, who hates the movement and all its works, but I think its success, and the overwhelming power and influence it now wields, is severely underestimated by its allies.)
I doubt you could have picked a worse example to make your point that contrarian takes are usually wrong than racial differences in IQ/intelligence.
Yeah, I'm not gonna get baited into getting rate-limited by downvotes that easily. All you need to know, if you care to, is that my values and goals differ drastically from most other people here.
EDIT: A mod rate-limited me anyway, perfectly illustrating the stark value difference: I consider that kind of abuse of power an atrocity.
a shallow clone
I don't know what that is. Genetically identical but not possessing my memories?
2. I wouldn't use the term "art" for that, no.
4. I don't have a constructive definition for what counts as an "artistic process," but I certainly wouldn't say the past holds no value for me: I like ancient sculptures and the like more than the next guy, and generally loathe iconoclasts and book burners. But if, say, the Library of Alexandria had books that had been printed instead of scribed by hand, I would not consider its burning any less of a loss.
5. No, I wouldn't. I have never used heroin, but its effects on others don't seem like the kind of thing I'd wish for myself permanently.
6–10 are easy. I'd consider the clone at least family.
I don't care to answer 1. and 3.
The "Disneyland without Children" short story? Yeah, I did. I'd read it before, and found it a nice fleshing-out of Bostrom's phrase. I do find it dystopic, but don't consider it illustrative of a likely future.
I don't think my disagreement is particularly insightful, but sure: if the painting is beautiful, I don't care if the artist is ugly or blind. With AI/ML image generation, we now have "beauty too cheap to meter" (Scott Alexander's phrase), and I don't see that as a bad thing. What "being human" means is something you construct for yourself, and a worry that it will somehow be lost or replaced is misguided.
Less high-mindedly, the (overwhelming majority of) human artists in power, writing novels, playing music, making movies, creating video games, translating anime/manga, have been telling me loudly and repeatedly for several years now that they despise me and everything I value, so fuck 'em (especially the anime/manga localizers); anything that democratizes their power, diminishes their status, throws open the gates they keep, is good, both for my community as a whole, as well as for me in particular.
Mukami-sama, the God of Atheism
Relevant smbc: link
There's another post in that series the first link missed. Look at the end of this one: link (sorry, low res image).
You might like this classic 4chan greentext: link.
What do most people gain from knowing this trivium? It's useful if you're signaling that you're on the side of "Science!" (which requires you to similarly "know" a lot of things that're a lot more dubious, and many things are are meaningless or outright false) but otherwise, unless you're an astrophysicist or similar, it makes no difference to you one way or the other.
The first few videos will necessarily be terrible, especially, hopefully, by the standards of the 47th video.
Suggestion: do them out of order.
"hey, your kids have a real chance of dying in the next decade"
Yes, every four years, if the good guys don't win the next (US) presidential election. Or if people don't switch to/away from nuclear power. Or they're killed by immigrants/cops. Or they die of a fentanyl overdose. Or in a school shooting. Or if the Iraqis/Russians/Chinese invade. Or if taxes are lowered/raised.
Perhaps telling people they or their children are going to die imminently isn't a standard tactic of "mere politics" where you are; you did say you're not American.
You'd think, but I wasn't been able to find such a thing despite looking pretty hard a few years ago; there might be a more recent AI approach to this though. A useful search term might be "audio to midi conversion". (Stem separation, for which Spleeter works well, might be a necessary preprocessing step.)
I'm not. I find your deontological murder exception ad hoc. My hypothetical hitman's moral foundations rest solidly on the notions, standard among "Effective Altruists," that one can value human lives in terms of money, that one can engage in tradeoffs involving lives, that the lives of people in Africa or wherever aren't worth more an order of magnitude less than those of people in, say, America, and that GiveWell's estimates of how much it costs to save a life are reasonable.
don't do work that is illegal, or that would be illegal if the public knew what you were really doing.
This rule only makes sense if you trust this "system for declaring profitable activities with negative externalities off limits" (ignoring that it pretends the law reflects the public's will far more than it actually does, and problems with manufactured consent) more than your own moral judgment, self-serving as it may be. Perhaps you believe the law is mostly just, or that your own moral reasoning is horribly flawed (side note: either both of these are true, or neither). I don't.
some kinds of harm (ex: murder) do not seem like the kind of thing you ought to be able to "cancel out" through donation, even if the donation clearly has larger benefits (ex: saves vastly many lives).
I disagree. An earning-to-give hitman is the epitome of the EA philosophy.
I hate the terms Concave and Convex in relation to functions.
Agreed.
the line (which should be considered to be the open side because integration makes the side below the line the solid side)
This is terrible: one pretty basic property you want in your definition describing the shape of functions is that it shouldn't change if you translate the function around.
Consider (for ). Pretty much the point of this definition is to be able to say it's the opposite kind as , but your choice wouldn't have that feature.
The "line at infinity" is a better choice for the imagined boundary. That's how we can think of parabolae as a kind of ellipse, for example.
if we called them decelerating (concave) or accelerating (convex) functions.
That'd be at least as confusing as the current terms for functions like or or .
You wrote "the world will always be exactly as it is." I don't see the difference.
until soon with SD3
I'll believe it when I see it. The man who said it would be an open release has just been fired stepped down as CEO.
Let the Lower Lights Be Burning
I had not heard this hymn before, but I love it! Thank you!
Hopefully, we'll have to write some new verses about polio soon.
And guinea worm.
Wtf? "God Wrote the Rocks" (which I love, and am grateful to the Solstice post a few months ago for pointing me to it) serves as excellent contrast to "Humans Wrote the Bible." These rewritten lyrics are just bizarre. The "book of earth", "book of night", "book of names"? "Humans write the book of truth" seems to be missing the whole point! It sounds good as long as you don't think about it, which strikes me as contrary to the whole "rationalism" schtick.
Also, the original lyrics are already pretty irreligious deism: this attempt at secularizing them is misguided in addition to being poorly executed.
there are many good reasons to think this whole ordeal will not work
Do any of these reasons not also apply to the cat in box?
Knowing this doesn't actually save the redwoods, necessarily, but it does make it far easier to be in a state of acceptance, because deep down nobody is actually your enemy.
This sounds like green-according-to-black.
My view of green is much more ecofascist-y.
That's not what "impervious" means: your view does not open itself up to falsification by logical argument or by experiment. Any argument against it would only address its internal consistency, which I think it fundamentally has; I was being only slightly sardonic when I said that was as good as truth.
You might like this fanfic (~1500 words): Mandragora.
Ah, honest-to-god supernaturalism! I didn't expect to find that here. It's a view impervious to reason or empiricism, but perfectly self-consistent, which makes it as good as true.
He's done something stronger than that: he has taken words that existed before, and replaced whatever meanings they may have had with his own. Elf, dwarf, orc, … these now indelibly bear his mark.
we don’t give a shit about morality. Instead, we care about social norms that we can use to shame other people
This reminds me of:
"Why didn't you tell him the truth? Were you afraid?"
"I'm not afraid. I chose not to tell him, because I anticipated negative consequences if I did so."
"What do you think 'fear' is, exactly?"
link.
What do you think 'morality' is, exactly, if not social norms?
wolf made the world to be a perfect reflection of wolf's will alone, and so wolf was free of any obligations to anyone, but then all of that got screwed up when a second being (coyote) came into existence and started doing its own thing and spreading around.
See also: Eru and Melkor.
You might prefer to link to Duncan Sabien's post here instead of Medium.
The desired quality for pupils is that they be indoctrinable, not stupid.
some South Pacific cultures viewed heterosexuality as sinful.
That sounds like Margaret Mead's work.
Would that even be a meaningful question? Thinking of it as a kind of PCA, there will be some axis, with a lot of correlations, and how you interpret that is up to you.
Lebanon tried this "balancing different factions" thing. Their system probably won't survive a census: the last one was in 1932.
And so began the US civil war.
Despite the name, the war between the states bears little resemblance to the civil wars in your other examples; since you picked Iraq, the invasion of Kuwait is a closer analogy than the post-election Sunni rebellion (though any war of conquest would be apt). But your description of the causes of secession is sound.
How expensive is the finetuning step relative to the pretraining (in terms of compute, data, labor, or anything else)?
I gather it'd be ~$1000 to "uncensor" a finetuned model, but as mentioned, this might be the first significant model released before finetuning, so I have no intuition for this. Two orders of magnitude more? Three?
Does something like the "I have been a good Bing. 😊" thing count? (More examples.)
I'd say that's a pretty striking illustration that under the surface of a helpful assistant (in the vein of Siri et al.) these things are weird, and the shoggoth is a good metaphor.
What was the old title? Something like "Misaligned AI doesn't kill people, misaligned people do"?
This new one sounds like it could make for a good slogan too: "To the average American, gun control is more lethal than guns."
Is a "celief" in something meaningfully different from an "alief" in its opposite?