Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ben Affleck's surprisingly comprehensive take on LLMs for video

twitter.com

Post reply on HN
Print view View original

51–60 of 187 posts

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#51

Post by haswell »

dartos wrote: post #22
haswell wrote: post #10

Earlier quoted context omitted.

It has yet to be seen if this can truly be generalized and if such an analogy holds. Go is a game constrained by an extremely narrow set of rules. Brute forcing potential solutions and arriving at something novel within such a constrained ruleset is an entirely different scenario than writing or film-making which occur in an almost incomprehensibly larger potential "solution" space. Perhaps the same thing will eventu…

Kind of tangential, but I don’t think you can brute force go. We’d run out of energy before it finished. Statistical models shortcut that search space. But I agree that just because a screwdriver is good at screws doesn’t mean its good at hammering a nail, even though both tools are made of metal.

Ahh good point. I was thinking about “machine plays itself repeatedly until it gets good” aspect of AlphaGo Zero and my brain jumped to brute forcing, but agree that’s a misnomer.

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#52

Post by MichaelZuo »

TheAceOfHearts wrote: post #44
bdndndndbve wrote: post #12

My favorite quote on this topic is "Why should I bother to read something you didn't bother to write?" Especially when you're talking about fiction and reading/watching for enjoyment, what does it matter if you can shit out 1000 hours of AI content? Maybe it's good to keep babies entertained? Studios have gotten so into the habit of treating "content" as a fungible commodity, but the fact is that even blockbuster mov…

AI tools will continue to get better, and they'll really shine when they can enable increasingly smaller teams of people to execute on their creative vision. Existing non-AI tools have already helped enable tiny teams to create media which is enjoyed by millions of people; the biggest and most recent example is how Source2 is used to animate Skibidi Toilet. But there's still room for increasing accessibility. The big…

When do estimate there to be quality AI generated content better than say the 90th percentile of Hollywood output?

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#53

Post by hnthrowaway6543 »

njtransit wrote: post #39
aetherson wrote: post #31

Earlier quoted context omitted.

This is the classic expression of the fallacy that the value of something is based on the cost to create it from the seller, not the benefit it brings to the buyer.

Additionally, there are lots of examples where cheaper production has produced an inferior product yet the difference in price causes the inferior product to usurp in superior product. Building materials exhibit this effect frequently: plaster vs. drywall, asphalt vs. slate, balloon framing vs structural masonry, etc. In media, TikTok exemplifies this effect. People watch fewer movies (expensive, high quality) and wa…

Saying movies are inherently a superior product to TikTok shorts is incredibly untrue. I would rather watch 90 minutes of the dumbest TikTok crap imaginable than sit through Madame Web again.

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#54

Post by xenospn »

kleiba wrote: post #29

The trope of the sci-fi robot has been so engrained into our cultural narrative that I think people constantly underestimate the creative capabilities of AI. People still think in terms of this dichotomy of math/science and art/creativity. But AI has already broken through so many boundaries of what anyone would have believed possible just a few years ago. I hear people complaining about uncanny valley stuff all the…

I use AI coding assistants every day and while they're helpful, they're nothing more than a glorified autocomplete.

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#56

Post by mchlbnnn »

Excellent art might be attributed less to genius and more to a selection process and creator-audience feedback loops.

We might be falling into a bit of survivorship bias when missing that excellent art is the product of a variety of processes, not just human inspiration.

Machines can probably replicate much of that process, for better or worse.

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#57

Post by coldtea »

aetherson wrote: post #31
bdndndndbve wrote: post #12

My favorite quote on this topic is "Why should I bother to read something you didn't bother to write?" Especially when you're talking about fiction and reading/watching for enjoyment, what does it matter if you can shit out 1000 hours of AI content? Maybe it's good to keep babies entertained? Studios have gotten so into the habit of treating "content" as a fungible commodity, but the fact is that even blockbuster mov…

This is the classic expression of the fallacy that the value of something is based on the cost to create it from the seller, not the benefit it brings to the buyer.

Art is the domain where "the cost to create it from the seller" matters.

Now, for those OK with slop, they can have it, but that's called content. Hollywood and SV (and most consumers) conflate the two all the time.

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#58

Post by taco_emoji »

TheAtomic wrote: post #4

Was it? He concludes that LLMs will never write Shakespeare, or create original work of that caliber, or animate an actor the way Ben Guinness could have done it. I'm paraphrasing but isn't this confirmaation bias from a guy now heavily vested in making movies? Move 37 was so astonishing the world champiion Go player, hardly known for theatrics, got up and left the table. It was considered an insanely "original" move…

Art is not a means of content generation. It is communication between consciousnesses. Why would an AI have anything interesting to tell me?

I feel absolutely insane when I read views like this, like I share the planet with body snatchers.

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#59

Post by coldtea »

TheAtomic wrote: post #4

Was it? He concludes that LLMs will never write Shakespeare, or create original work of that caliber, or animate an actor the way Ben Guinness could have done it. I'm paraphrasing but isn't this confirmaation bias from a guy now heavily vested in making movies? Move 37 was so astonishing the world champiion Go player, hardly known for theatrics, got up and left the table. It was considered an insanely "original" move…

>In every field we don't consider deeply "human," Ai has jumped light years ahead, in astonishing ways

Didn't you answered yourself?

Re: Ben Affleck's surprisingly comprehensive take on LLMs for video

#60

Post by sumtechguy »

moralestapia wrote: post #16
moogly wrote: post #11

Earlier quoted context omitted.

> He even contradicts his whole premise of "actors will always be necessary" with his example of a custom generated series episode. He said it would be "janky and weird" and likened it to a remix.

People who think generating language is very much different from generating audio/video are in for a ride, which I think is the flaw with his reasoning. (Also, yeah, there's a very strong interest for him in saying actors will not be replaced, but let's assume he's being frank here). All the "AI personalities" that are already popping up here and there are already decent enough, again, it will only take a couple year…

If what the churn mills are any indicator the coherence is already starting to get there. Saw one this weekend where it was star trek set in the 1940s sci-fi vibe. The coherence was a bit off. But way better than it was last year with the same kind of videos. Even from scene to scene it is getting better. In scene the warping is starting to go away. Add in the ability to write a story using chatgpt (or something like it) then add in the 'ai voices' and you can pick and choose who/what you want.

Movie making is about to radically shift to an extremely lower cost model to produce. Just not quite yet.

I can already ask chatgpt to 'write me a SCP story about a couch found in an abandoned factory that is keter class' and it will pump out something mildly entertaining in that style. It is not that big of a leap of logic to say 'good enough' will win out. Even now you can kinda pull it off with a small amount of work and stitching several tools together. There is no reason to think it will not be automated together.

Post reply on HN