4 ms·

This will age like a dead fish out in the Sun. He even contradicts his whole premise of "actors will always be necessary" with his example of a custom generate

This will age like a dead fish out in the Sun. He even contradicts his whole premise of "actors will always be necessary" with his example of a custom generated series episode. Also, >implying the non-AI content we have nowadays is highly creative and unique *facepalm*

sottol 2y ago
How soon though? 2 years? 10? 50? He starts by saying this won't happen soon, _maybe_ ever.
- moralestapia 2y ago
  I'd bet that in two years you could do something equivalent to that "make me a new episode" use case. I already see really convincing deep fakes on Facebook, etc., it's not an impossible problem anymore, it's just a matter of increasing their quality.
  - dham 2y ago
    Convincing deepfakes is a strong word. There are also people out there who think modern CG looks good, too. There will always be people who are ultra-sensitive to things like this. So convincing might take 2x as long for some vs others.
    - ruthmarx 2y ago
      > There will always be people who are ultra-sensitive to things like this. There are always people that claim to be...
      - dham 2y ago
        True, but it doesn't change the fact that some percentage of the population is.
        ruthmarx 2y ago
        I'm kind of skeptical, just because I've seen people claiming things are CGI when they were not, or claiming the CGI was crap when it was fine. Reminds me of photographs asking people if photos were photoshopped or not and people generally guessed wrong about half the time.
    - sumtechguy 2y ago
      There is a musical artist on YT talking about this. He was sitting there listening to AI generated music. He said he could not tell and thought it was decent. His kid and wife however could and aske why he was listing to 'ai music'. That is a professional musician who couldn't tell. My wife can hear something she can not describe that makes her say 'it sucks'. Meanwhile I listen and think it is OK. It is like these things have tweaked that audio/visual hallucinations people see and hear like yannie/laural or the gold/blue dress. The corridor crew noticed something about AI generated pictures. They all seem to have similar lighting and chroma. So they have this 'look' to them (once you see it). That is because the pictures start off as a random noise picture and coalesce into a picture. So the lighting stays the same on just about every pixel.
  - no_wizard 2y ago
    Is there any deep fake video that is possibly convincing that is 100% generated synthetically? All the deep fakes I've seen that don't look incredibly weird at first glance essentially re-use clips (like Back To The Future[0]) and they replacement an element, like a face, but the rest of the footage already exists. If you look closely though, particularly at the mouth of the video I linked, something isn't quite natural about it still. Same deal with this Taylor Swift fake[1] [0]: https://www.youtube.com/watch?v=8OJnkJqkyio https://www.youtube.com/watch?v=8OJnkJqkyio [1]: https://x.com/McAfee/status/1745226438641602866 https://x.com/McAfee/status/1745226438641602866
  - xenospn 2y ago
    Getting to 90% is easy. The last 10% can take 50 years. Don't forget we're the product of evolution and subconsciously look for things current technology will not be able to replicate.
moogly 2y ago
> 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.
- moralestapia 2y ago
  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 years until their quality is on par with meatspace video. Then it's gg for actors (and many other professions).
  - dartos 2y ago
    > 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. How could anyone think they are similar? You don’t just pump video data into an LLM training pipeline and get video models. A lot of work goes into both. It’s non trivial and only barely related. It’s only recently that transformer models are being used for video.
  - sumtechguy 2y ago
    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.
aeturnum 2y ago
You're mis-reading what he says. People create new stuff, AI (so far) remixes old stuff. Down the line you will be able to order a remix of succession and AI should be able to do that for you but it won't be able to add any interesting additional spin - it's just gonna make it to order (obviously a big deal! which he says). >implying the non-AI content we have nowadays is highly creative and unique You just inserted this, he doesn't say or imply this - he asserted that all of the new ideas (so far) have been generated by humans. Sure lots of it is slop - but all of AI content is slop. That's his point: the new, creative stuff is still coming from humans - even if they are using "AI" tools to do it.
- moralestapia 2y ago
  ??? Show me a movie in the last decade where the premise of it couldn't be generated by some AI prompt. (I'll even make it easier by stating that said AI prompt can be generic and to some degree trivial; i.e. not the premise explicitly laid out in the prompt).
  - ricoxicano 2y ago
    ??? show me an AI prompt that can generate a movie premise in the last decade where the underlying model hasn't already been trained on that movie's plot.
    - danielmarkbruce 2y ago
      show me a writer that hasn't been trained on a lot of movie plots before them.
  - jjj123 2y ago
    Believe it or not, but there’s a lot more that goes into filmmaking beyond the premise. Parasite, The Substance, Uncut Gems, mother!, Krisha, Tangerine, and The Lighthouse are some movies from the last decade that are novel in ways beyond their premises. No generic prompts to an LLM could create those movies, even if it could spit out a vague 2 sentence premise that matches them. Note that I’m not saying it could never happen, but LLMs as we have them in 2024 produce regurgitated slop, and require significant creative input from the prompter to make anything artistically interesting.

sottol 2y ago

How soon though? 2 years? 10? 50? He starts by saying this won't happen soon, _maybe_ ever.

moralestapia 2y ago
I'd bet that in two years you could do something equivalent to that "make me a new episode" use case. I already see really convincing deep fakes on Facebook, etc., it's not an impossible problem anymore, it's just a matter of increasing their quality.
- dham 2y ago
  Convincing deepfakes is a strong word. There are also people out there who think modern CG looks good, too. There will always be people who are ultra-sensitive to things like this. So convincing might take 2x as long for some vs others.
  - ruthmarx 2y ago
    > There will always be people who are ultra-sensitive to things like this. There are always people that claim to be...
    - dham 2y ago
      True, but it doesn't change the fact that some percentage of the population is.
      - ruthmarx 2y ago
        I'm kind of skeptical, just because I've seen people claiming things are CGI when they were not, or claiming the CGI was crap when it was fine. Reminds me of photographs asking people if photos were photoshopped or not and people generally guessed wrong about half the time.
  - sumtechguy 2y ago
    There is a musical artist on YT talking about this. He was sitting there listening to AI generated music. He said he could not tell and thought it was decent. His kid and wife however could and aske why he was listing to 'ai music'. That is a professional musician who couldn't tell. My wife can hear something she can not describe that makes her say 'it sucks'. Meanwhile I listen and think it is OK. It is like these things have tweaked that audio/visual hallucinations people see and hear like yannie/laural or the gold/blue dress. The corridor crew noticed something about AI generated pictures. They all seem to have similar lighting and chroma. So they have this 'look' to them (once you see it). That is because the pictures start off as a random noise picture and coalesce into a picture. So the lighting stays the same on just about every pixel.
- no_wizard 2y ago
  Is there any deep fake video that is possibly convincing that is 100% generated synthetically? All the deep fakes I've seen that don't look incredibly weird at first glance essentially re-use clips (like Back To The Future[0]) and they replacement an element, like a face, but the rest of the footage already exists. If you look closely though, particularly at the mouth of the video I linked, something isn't quite natural about it still. Same deal with this Taylor Swift fake[1] [0]: https://www.youtube.com/watch?v=8OJnkJqkyio https://www.youtube.com/watch?v=8OJnkJqkyio [1]: https://x.com/McAfee/status/1745226438641602866 https://x.com/McAfee/status/1745226438641602866
- xenospn 2y ago
  Getting to 90% is easy. The last 10% can take 50 years. Don't forget we're the product of evolution and subconsciously look for things current technology will not be able to replicate.

moogly 2y ago

> 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.

moralestapia 2y ago
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 years until their quality is on par with meatspace video. Then it's gg for actors (and many other professions).
- dartos 2y ago
  > 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. How could anyone think they are similar? You don’t just pump video data into an LLM training pipeline and get video models. A lot of work goes into both. It’s non trivial and only barely related. It’s only recently that transformer models are being used for video.
- sumtechguy 2y ago
  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.

aeturnum 2y ago

You're mis-reading what he says. People create new stuff, AI (so far) remixes old stuff. Down the line you will be able to order a remix of succession and AI should be able to do that for you but it won't be able to add any interesting additional spin - it's just gonna make it to order (obviously a big deal! which he says). >implying the non-AI content we have nowadays is highly creative and unique You just inserted this, he doesn't say or imply this - he asserted that all of the new ideas (so far) have been generated by humans. Sure lots of it is slop - but all of AI content is slop. That's his point: the new, creative stuff is still coming from humans - even if they are using "AI" tools to do it.

moralestapia 2y ago
??? Show me a movie in the last decade where the premise of it couldn't be generated by some AI prompt. (I'll even make it easier by stating that said AI prompt can be generic and to some degree trivial; i.e. not the premise explicitly laid out in the prompt).
- ricoxicano 2y ago
  ??? show me an AI prompt that can generate a movie premise in the last decade where the underlying model hasn't already been trained on that movie's plot.
  - danielmarkbruce 2y ago
    show me a writer that hasn't been trained on a lot of movie plots before them.
- jjj123 2y ago
  Believe it or not, but there’s a lot more that goes into filmmaking beyond the premise. Parasite, The Substance, Uncut Gems, mother!, Krisha, Tangerine, and The Lighthouse are some movies from the last decade that are novel in ways beyond their premises. No generic prompts to an LLM could create those movies, even if it could spit out a vague 2 sentence premise that matches them. Note that I’m not saying it could never happen, but LLMs as we have them in 2024 produce regurgitated slop, and require significant creative input from the prompter to make anything artistically interesting.