I think asking some basic troubleshooting questions is helpful.
Is your product worth paying for? Is it even worth someone’s time to look at it or download it for free?
Is it the type of thing that you yourself have paid for in the past? (E.g. if you’re making infographics about cooking, ask yourself, when’s the last time you gave someone money because of their cooking infographics). Would you pay for your own product?
If you can’t imagine a scenario where you would pay for your own product, then why would anyone else?
Naturally, the first step for AI mathematics would be generating proofs of theorems that humans have already deemed interesting, rather than the AI generating entirely new research directions all by itself.
Certainly it’s an interesting problem to think about how one would formalize the notion of interesting/important mathematics. But it’s not necessary to solve that problem in order for AI to start having a big impact on how mathematics is done.
>Each time this happens, the AGI pessimists raise the bar (a little) for what constitutes AGI.
Let’s say I have a coworker. We write code together, but we also hang out outside of work sometimes, maybe go for a couple rounds of golf, maybe discuss our shared passion for indie rock, he tells me his opinions on current news events or what’s going on with his family, all the ordinary sorts of things that people do.
When we get to the point where I have to reasonably wonder if my coworker is a biological human, or an AI masquerading as a human, then we have undoubtedly achieved AGI.
Not that we actually have to flood society with an army of cybernetic humans, of course. It’s just that AGI would enable such a thing, in principle.
Do you think you’ll see true AGI in your lifetime? I certainly don’t think I’ll see it in mine.
No other domain of the sciences is “complete” yet. There are still unanswered questions in physics, biology, neuroscience, every field. Why are people so sure that AI will be the first field to be “completed”, and in such an astoundingly short amount of time on a historical scale?
I agree. People will be quick to say that that’s just your nostalgia talking, but taken to an extreme, that argument would entail that culture and patterns of behavior never change, which is equally implausible.
Well, no, I wouldn’t say that my post was about “intention”. I could imagine, in principle, someone saying “hey GPT-3, write me a philosophical masterpiece” or “write me a hit novel” and it does. No intentionality required. It’s just that in actuality, it can’t do that yet.
Between this and DALL-E 2, I’ve been ruminating on David Hilbert’s old quip about a student: “For a mathematician, he did not have enough imagination. But now he is a poet, and everything is fine.”
As someone with experience in both STEM and the arts, I don’t feel particularly bad about saying that some “creative” pursuits are more creative than others, and my rough thesis is that the ones on the higher end of the scale are the ones that will take longer to fall to AI.
Representational painting is, at the end of the day, a somewhat mechanical problem, so it’s not too surprising that AI is rapidly catching up to humans there. “Draw me a cat eating ice cream on the beach” is a fairly well-defined request with a clear range of acceptable solutions.
I think the creativity exhibited in a long-form written work like say, Plato’s “Republic”, is of a much higher degree than your average painting or short poem. Not only is it an aesthetically beautiful work, but it contains a great deal of novel (for its time) intellectual content as well, filtered through the personal perspective of the author, and organized in an elaborate logical structure. When it comes to these types of large-scale creative works, AI hasn’t even begun to tackle the problem yet.
I don’t really disagree. I’m not a luddite and I don’t advocate for turning off the internet. Computers are certainly here to stay. It’s an extremely complex issue, and I don’t have all the answers, or even know how to phrase all the questions.
I do think society needs to take a proactive role in deciding how it wants to interact with technology though. There’s a certain laissez faire, almost defeatist attitude that you see from a lot of the tech crowd, that goes something like “technology will do what it does, and it will change our lives how it sees fit, and we are powerless to stop it.” But if that was the case, we couldn’t have gun control laws, or environmental protection laws, or restrictions on nuclear technology. Technology may continue to develop, but it’s still up to us how we choose to use it.
Sorry, that may have been poorly worded. I wanted to preempt the objection of “well, you say you don’t want to be dependent on smartphones, but then you’ll just be dependent on some other type of computer”. I wanted to make it clear that the problem is about rethinking our relationship with computing in general, not just with smartphones.
I simply won’t use any service that requires a phone and doesn’t allow other options. I am opposed to a future in which phones are a necessity of life rather than merely a convenience.
And to the people who say “but desktops/laptops are already a necessity of life” - yes, and that’s a problem. We need to be actively thinking of ways to roll things back, rather than allowing technology to become more and more integrated into life.
>Honestly, I'd assume that pretty much any of these flaws are more fixable than you'd expect.
The first 90% of the project takes 90% of the time, the last 10% of the project takes the other 90% of the time.
I don't think it's wise to speculate at this point on what problems are "actually easy" or "actually hard", or how long it'll take to achieve certain milestones. I don't think that even the people at the forefront of this research have any way of knowing for sure.
DALL-E 2 is very impressive. So impressive that many people have wondered if all commercial artists will be out of a job as soon as DALL-E 2 is made publicly available. These limitations form a strong argument that DALL-E 2 is not ready to completely replace human artists just yet. This is particularly relevant if you're an artist, or if you're in the market for one.
>Instead, we're only interested in finding the logically valid theorems which serve the above purpose of explaining the patterns in the world concisely.
That's just empirically false though. Take the "perfectoid spaces" mentioned in the article. Do you think mathematicians are interested in them because they "explain the patterns in the world concisely"? No, they're interested in them because of their mathematical significance, and their applications to problems within mathematics. Mathematicians are not physicists.
I mean, I know this conversation is all about railing against the snobbish "you just don't understand fine art" attitude, but I find the stubborn "he really needs to reign it in and not be so crazy" attitude to be unfortunate too. It's good for artists to experiment. A lot of those experiments can be very fun and enjoyable for the readers, if you're willing to give them a chance.
>You'd essentially be stating that a whole domain of tradition, practice, and procedure which organically grows and evolves, and I might add, in almost logical progressions at times, was ousted by the whims of one foppish professor who gamed everyone into liking something he liked simply because he liked it and was impassioned enough about it.
I mean, take this argument and apply it to something like theology. "Are you really going to say that an entire tradition, one which has produced innumerable great thinkers and has proceeded on a logical progression towards truth, is entirely mistaken in its most fundamental assumptions?" It turns out that, yeah, I would say that. Sometimes people make mistake. Sometimes lots of people make lots of mistakes and the mistakes go on for thousands of years.
>For instance, you say you have never heard a convincing argument as to why Pynchon would be considered worthy of canon status--well, I'm not going to be so absurd as to claim he'll still be there in 200 years, but if you have knowledge of literary craft the reasons why he's there now are pretty clear--his maximalism is both well crafted and unique and his style is a turn away from the still dominant style of american literature (Hemmingway based minimalism) which is positively refreshing
But this is exactly the issue. What does it mean for writing to be "well crafted", what does it mean for writing to be "positively refreshing"? If we can't give rigorous, verifiable definitions for these concepts, then we're just saying "he's good because he's good".
>It's because he utilizes traditional structures and devices in a unique way
Originality has at least the hope of being a more objective metric, although originality is clearly a very slippery concept. If I take a famous novel and change one word, the result may be a work that has never existed before, but that's not originality. If I use a computer to generate a completely random image, then again, that image may have never existed before, but that's not originality. So it's very hard to define. But I at least see the hope of a project there.
>I can understand where his pieces fit into the narrative of painting history, what they challenge, what they change, and ultimately how unique his forms are and what they communicate within this context. If I dislike it, if I find it shouldn't be considered art--well I have to argue it from this perspective, from within the game of homo sapiens art history.
I can certainly appreciate and enjoy playing games, since some games are very beautiful. But this "game", the game of "make an original contribution to art history and get academics to talk about it", seems to have no rules! What good is a game if no one can tell you the rules? The judges of the game can gesture towards criteria like "originality" that might conceivably give you some guidelines on how to play, but no one can definitively prove that one person deserved to win and another didn't.
>artistic forms exist because there are elements of these traditions a large number of people can generally agree they appreciate
Yes, I agree. But if we tried to submit the works of the canon to an analysis of this kind, to see how many of them contain "artistic forms that people agree they can appreciate", how many of them would survive? Some would, I'm sure. I think Shakespeare and Homer, for example, are still legitimately appealing to people today, if they can work past the archaic language. But exactly how many works of the canon would survive this analysis? Does Schoenburg's music have "forms that people agree they can appreciate"? If not, then what does that say about the validity of Schoenburg's status as a great canonical composer? (I happen to enjoy a lot of Schoenburg's music, but I don't think someone is stupid if they don't).
>Thus the cannon isn't a good judge of anything other than what your precursors believed should be appreciated.
All that effort, and this is the conclusion we're left with!
>It's essentially the historical development of a shared value judgement, or a shared human prejudice. So of course you can repudiate the whole thing. But at that point you are no longer even engaging in that art form--or at best you are engaging with blinders on, and any aesthetic mastery you manage to pull off is largely lucky and unconscious. You are starting from a different base. you are playing your own game. Thus you shouldn't be too upset when other people don't appreciate what you do, or your work isn't considered interesting. You're not even speaking their language.
Ok, this is a really interesting paragraph. You say it's the development of a "shared value judgement", but who, exactly, shares this value judgement? We have, on the one hand, a comparatively small community of academics who share the value judgement that Proust and Melville are wonderfully nourishing authors who deserve to be read again and again by new generations, and on the other hand, we have hundreds of millions of people who would just as soon throw Proust and Melville in the trash so they could go watch the latest Marvel movie or listen to the newest Justin Bieber song. Taking your comment about the "historical development of value judgement" seriously, what can we say by looking at this concrete historical moment? What can we say about what that value judgement has become? What authority can the university canon possibly have in the face of this sheer numerical onslaught? You say that "you shouldn't be too upset when other people don't appreciate what you do, or your work isn't considered interesting. You're not even speaking their language," but who exactly are the artists who are being appreciated these days? Who is speaking the language that most people find congenial? It's certainly not the academics or the purveyors of "high culture", the defenders of the "artistic tradition". MoMA exhibits of great theoretical sophistication are laughed at while Jay-Z packs stadiums. Your warning applies much more to all those who would play that slippery, ephemeral game known as "art history", rather than those who would simply "copy what they like".
All that being said, I do think the notion of a "canon" could be saved, but it has to be grounded in genuine, verifiable scholarship. With Shakespeare, for example, we could do research on his influence on the English language, his influence on other artists, his continuing popular appeal, etc, and come to the conclusion that his plays constitute a major accomplishment. But this game of "aesthetic criticism", or declaring this or that work to be beautiful or enlightening or whatever, that doesn't need to be done in universities. People can do that on their own time. Like I said above, a game where you can't even know the rules can't hold your attention for very long. You're better off playing Go or doing math. At least there you can know that you're winning.
Is your product worth paying for? Is it even worth someone’s time to look at it or download it for free?
Is it the type of thing that you yourself have paid for in the past? (E.g. if you’re making infographics about cooking, ask yourself, when’s the last time you gave someone money because of their cooking infographics). Would you pay for your own product?
If you can’t imagine a scenario where you would pay for your own product, then why would anyone else?