It's only going to get better, faster, cheaper, easier.[a]
Sooner than anyone could have expected, we'll be able to ask the machines: "Turn this book into a two-hour movie with the likeness of [your favorite actor/actress] in the lead role."
Sooner than anyone could have expected, we'll be able to have immersive VR experiences that are crafted to each person.
Sooner than anyone could have expected, we won't be able to identify deepfakes anymore.
Are we only a few years away from one person/small group made movies where the dialog, acting, location and special effects can be tweaked endlessly for a relatively low cost. If I was a studio exec I'd be worried.
I don’t know, for a car the last 10% has a direct relation with "people die" that is obvious to everyone. With the movie made in anyone basement, the risks are far less likely to create such a vivid perception of dramatic end result.
Not that cyber-bullying and usurpation schemes escalating a whole new level being less of a concern in the aftermaths, to be clear.
Less about risk parallels and more about control parallels. The last 10% of fine grained control over a system is hard. Like every time I’ve done text to image prompting and it gives me a great starting point, but cannot get certain details i want, no matter how i ask.
The average person spends 9-11 hours per day consuming media depending on what source you look at. When people are playing games or browsing social media at the same time that they have the latest Netflix show on their TV, you can't tell me that this is really valuable time spent to deepen one's understanding of the human experience; it's a replacement for the human experience.
Most people will not notice if the soundtrack to a new TV show is made by a 5 word AI prompt of "exciting build-up suspense scene music" while they're playing pouring money into their mobile gacha game to get the "cute girl, anime, {color} {outfit}" prompt picture that is SSS rank.
You or I might not care for AI slop, but it's a lot cheaper to produce for Netflix or Zinga or Spotify or whatever, and if they go this route, they don't have to pay for writers, actors, illustrators, songwriters, or licensing for someone else's product. They'll just put their own AI content on autoplay after what you're currently watching, and hope most people don't care enough to stop it and choose something else.
If we judge from AI writing, we can extrapolate what an AI movie would be like. I cannot imagine reading an AI book. It would look and smell like a book but nothing of value or new insight would be inside. Michael Bay might be very interested.
You look back at old movies, and on a technical level they really aren’t as good as contemporary trash productions. But they knew how to weave the camera and a script into something amazing back then even if they didn’t have resolve and aftereffects to polish every shot. A good script writer, editor and cinematographer have a huge impact on the quality of a movie. But these roles are only a small portion of the operating budget of a movie. Filming every single scene is an exhausting undertaking and this constitutes the bulk of a movie production’s budget. If you can get good quality footage without leaving your garage then you can have a small team make a great movie. Maybe not the extent where you simply click a button but to the level that you would launch straight to a streaming service.
Yes, AI will probably fail miserably for a while at least, at making the sort of well written artistic, clever movie that nobody watches. The only ones that need to be worried are the studios making churning out massive blockbusters…
If you look at the majority of their catalogue these days, they really aren't trying to squeeze that last 10% out of the movie quality these days anyways, so I doubt it will matter.
to be fair, that's exactly what the asset class EXISTS for - betting on huge outcomes, no matter the odds. People misunderstand that due to how much of tech is "VC funded" when building stuff that would fare better as a bootstrapped company (or funded by other means)
You don't get "The Green Mile" from this, because it's a tool. You get "The Green Mile" with artistic vision. The tool has to be told what to do. But now a director can shoot a film with actors who don't match the physical description of a character in a story, and then correct their race/gender/figure/whatever with AI. That probably means they save money on casting. A director can shoot a scene inside a blank set and turn it into a palace with AI. That saves money from shooting on location or saves money from having to pay for expensive sets.
So now a director with a limited budget but with a good vision and understanding of the tools available has a better chance to realize their vision. There will be tons of crap put out by this tool as well. But I think/hope that at least one person uses it well.
But because it will make shooting a movie more accessible to people with limited budgets, the movie studios, who literally gatekeep access to their sets and moviemaking equipment, are going to have a smaller moat. The distribution channels will still need to select good films to show in theaters, TV, and streaming, but the industry will probably be changing in a few years if this development keeps pace.
This is the best answer I've seen, but I think what was demonstrated is miles away from this. A lot would need to be able to be specified (and honored) from the prompts, far more than any examples have demonstrated.
I'm not against tools for directors, but the thing is, directors tell actors things and get results. Directors hire cinematographers and work with them to get the shots they want. Etc. How does that happen here?
Also, as someone else mentioned, there is the general problem that heavily CG movies tend to look... fake and uncompelling. The real world is somehow just realer than CG. So that also has to be factored into this.
I think it starts simple. Have you ever been watching a movie or tv show, and it shows the people walking up to the helicopter or Lamborghini and then cut to "they've arrived at their destination no transportation in sight"?
It will start out with more believable green screen backgrounds and b roll. Used judiciously, it will improve immersion and cost <$10 instead of thousands. The actors and normal shots will still be the focus, but the elements that make things more believable will be cheaper to add.
Have you ever noticed that explosions look good? Even in hobby films? At some point it became easy to add a surprisingly good looking explosion in post. The same thing will happen here, but for an increasing amount of stuff.
Interesting that you pick that example in particular. Due to the sheer depth of behinds the scenes takes HBO has provided for Game of Thrones and House of the Dragon, it seems to be the consensus view among effects folks that CG fire and explosions are nearly impossible to get right and real fire is still the way to go.
That I could believe, although... there is quite a bit of commentary from film buffs that lots of the stuff done in post doesn't quite look right, compared to older films.
Which doesn't mean it won't keep happening (economics), but it doesn't necessarily mean any improvement in movie quality.
My guess is the art form will evolve. When YouTube started, some people thought it would not be able to compete with heavily produced video content. Instead, YouTube spawned a different type of "movie". It was short-form, filmed on phone cameras, lightly scripted, etc. The medium changed the content. I suspect we will see new genres of video content show up once this tech is widely available.
The first real movies made 100 years ago looked like something someone today could put together in their garage on a shoestring budget. AI-generated movies have existed for just two years, and are only going to get better. This is bleeding edge research, and I haven't seen any sign yet of AI models hitting a quality ceiling.
Shameless plug: I just created a short AI film (1) and tried to tell an actual story and trigger emotions. I spent countless hours crafting the script, choosing the shots, refining my prompts, generating images, animating images, generating music, sounds, and so on... For me AI tools are just that - tools. True, you have to yield them some "control", but at the end of the day you are still the one guiding and directing them.
Similarily, a film director "just" gives guidance to a bunch of people: actors, camera operator, etc. Do you consider the movie is his creation, even if he didn't directly perform any action?
A photographer just has to push a button and the camera captures an image. Is the output still considered his creation?
Yes and Yes, so I think we should consider the same with AI assisted art forms.
Maybe the real topic is the level of depth and sophistication in the art (just like the difference between your iPhone pictures and a professional photographer's) but in my opinion this is orthogonal to it being human or AI generated.
To be honest so far we have mostly seen AI video demos which were indeed quite uninteresting and shallow, but now filmmakers are busy learning how to harness these tools, so my prediction is that in no time you will see high quality and captivating AI generated films.
this is an excellent example that despite all the technical limitations (the ugly image artifacts, the lack of exact image consistency, etc), it's _already possible_ to create something that connects. The "format purists" currently dismiss AI tooling the same way they used to dismiss computer graphics animation back when Toy Story 1 came out.
I'm in agreement with scudsworth here, but i have a little more nuance, i think. I know how long this took, and how many compromises were made. The only reason this works, at all, is because humans have a massive list of cultural memes and tropes that shorthand "experience" for us. it has the "AI can only generate 2-5 seconds of video before it goes completely off the rails" vibe; which allows it to fit in with the ADHD nature of most video production of the last 30 years - something a lot of people do not like. For an example where this was jarring in the video, when they're drawing or painting, you cut the scene slightly too late, you can see the AI was about to do some wild nonsense.
What happened to the mom? Why does the kid get older and younger looking? why does the city flicker in the beginning? which kid is his in the ballet performance? why do they randomly have "lazy eye"? i could keep going but i think we all get my point.
I can intuit the tropes used by the AI to convey meaning, and i'd be willing to list them all with relevant links for the paltry sum of $50. Be warned, it will be a very large list. Tropes and "memes" are doing 100% of the heavy lifting of this "art".
Sorry, human. As someone who stopped creating art on a daily basis due to market dilution (read: it's too hard to build a fanbase that i care about), i am very critical of most "art" produced anyhow.
sure it is. that's my critical evaluation of this work. if you liked it, i highly recommend the hallmark channel, lifetime, family channel originals, the netflix straight-to-vod swimlane, and a frontal lobotomy.
I, uh, gave some more nuance because i had some free time as a sibling comment to this. I hope we don't get downvoted because someone has to call a spade a spade.
good comment, haha. agree with those points and would add, since im thinking about it again now, that the entire work feels like a fairly (deeply) shallow riff on the opening sequence in pixar's "up". but of course with no stakes or emotional impact whatsoever.
> How is any of this going to lead to meaningful art?
Nearly all the movies that go to theaters aren't "meaningful art". Not only that but what's meaningful to you isn't necessarily what's meaningful to others.
If someone can get their own personal "Godzilla VS The Iron Giant" crossover made into a feature-length film it will be meaningful to them.
They are art compared to getting uncontrolled choices. Who decides what the actors look like? How they move? What emotions their faces are to convey? How the blocking works for a scene? What the color scheme is for the movie? How each shot is taken? Etc.
There is a vast difference between a formulaic hollywood movie and some guy with a camera. If I say "Godzilla vs. The Iron Giant" what is the plot? Who is the good guy? Why does the conflict take place?
AI will come up with something. Will it be compelling even to the audience of one?
As a toy, maybe. As an artistic experience... not convinced.
This is a common perspective among people that don't realize how much goes into making a movie. That stuff informs which movies get approved and it certainly can inform broader script changes, casting changes, and in some cases editing decisions, but there's a UNIVERSE of other artistic decisions that need to be made. Implying that the people involved are mere technicians implementing a marketing strategy is exponentially more reductive than saying developers and designers aren't relevant to making software because marketing surveys dictate the feature development timeline. A developer's input is far more fungible than an artist working on a feature film.
I assure you, they don't do surveys on the punchiness and strategy used by foley artists; the slope and toe of the film stock chosen for cut scenes by the DP or that those cut scenes should be shot like cut scenes instead of dream sequences; the kind of cars they use; how energetic the explosions are; clothing selection and how the costumes change situationally or throughout the film; indescribably nuanced changes in the actor's delivery; what fonts go on the signs; which props they use in all of the sets and the strategies they use to weather things; what specific locations they shoot at within an area and which direction they point the camera, how the grading might change the mood and imply thematic connections, subtle symbolism used, the specifics of camera movements, focus, and depth of field, and then there's the deeeep world of lighting... All of those things and a million others are contributions from individual artists contributing their own art in one big collaborative project.
> Implying that the people involved are mere technicians implementing a marketing strategy
Well no. Instead, I am implying that they are as much of a "technician" so someone who is putting in a huge amount of effort into making AI videos.
If you want to say that it is perfectly possible for someone to put a high amount of vision and make a large amount of creative decisions into AI videos, then I agree.
> All of those things and a million others are contributions
Yes I agree that there can be a million other contributions to making AI videos. Glad you agree too.
Some of it is done that way, but by no means all of it. You can easily see the differences, because, say, Wes Anderson movies are not the same as Martin Scorsese movies.
If it were really all just market decisions, directors would have no influence. This is not remotely the case. Nor are they paid as though that were true.
> Nearly all the movies that go to theaters aren't "meaningful art".
No but what they are is expensive, flashy, impressive productions which is the only reason people are comfortable paying upwards of $25 each to see them. And there's no way in the world that an AI movie is going to come anywhere close to the production quality of Godzilla vs Kong.
And like, yeah, their example videos at the posted link are impressive. How many attempts did those take? Are they going to be able to maintain continuity of a character's appearance from one shot to the next to form a coherent visual structure? How long can these shots be before the AI starts tripping over itself and forgetting how arms work?
My suspicion is that, if AI moviemaking actually becomes common, there will be a younger generation of folks who will grow up on it and become used to its peculiarities.
We will be the old ones going "back in my day, you had to actually shoot movies on a camera! And background objects had perfect continuity!" And they will roll their eyes at us and retort that nobody pays attention to background objects anyway.
My suspicion (and fear) is that poor members of the younger generation are going to grow up reading AI kids books and watching AI TV shows, and playing AI generated iPad games, and be less literate, less experienced, less rounded and interesting people as a result. This is already kind of a problem where under-served kids access less, experience less, and are able to do less and I see AI doing nothing but absolutely slamming the gas on that process and causing already under-served kids to be even more under-served. That human created art will be yet another luxury only afforded to the children of the advantaged classes.
And maybe they won't have a problem with it, like you say, maybe that'll just be their "normal" but that seems so fucking sad to me.
If poor kids of the future grow up reading AI book-slop instead of classic books that's going to be due to complicated factors of culture and habit rather than economic necessity. Most of the traditional Western canon of "great literature" is already in the public domain, available for free.
It gives anyone with a participating-system library card free electronic access to books and magazines. And it's unlikely that librarians themselves will be adding AI book-slop to the title selection.
> If poor kids of the future grow up reading AI book-slop instead of classic books that's going to be due to complicated factors of culture and habit rather than economic necessity.
To be clear, I'm not talking great literature. I'm talking Clifford the Big Red Dog type stuff.
That said I still have a number of problems with this assertion:
It will absolutely be down, in part, to economic necessity. Amazon's platform is already dealing with a glut of shitty AI books and the key way they get ahead in rankings is being cheaper than human-created alternatives, and they can be cheaper because having an AI slop something out is way less expensive and time consuming than someone writing/illustrating a kid's book.
Moreover, our economy runs on the notion that the easier something is to do, the more likely people are to do it at scale, and vetting your kids media is hard and annoying as a parent at the best of times: if you come home from working your second job and are ready to collapse, are you going to prepare a nutritious meal for your child and set them up with insightful, interesting media? No you're going to heat up chicken nuggets and put them in front of the iPad. That's not good, but like, what do you expect poor parents to do here? Invent more time in the day so they can better raise their child while they're in the societal fuckbarrel?
And yes, before it goes into that direction, yes this is all down to the choices of these parents, both to have children they don't really have the resources to raise (though recent changes to US law complicates that choice but that's a whole other can of worms) and them not taking the time to do it and all the rest, yes, all of these parents could and arguably should be making better choices. But ALSO, I do not see how it is a positive for our society to let people be fucked over like this constantly. What do we GAIN from this? As far as I can tell, the only people who gain anything from the exhausted-lower-classes-industrial-complex are the same rich assholes who gain from everything else being terrible, and I dunno, maybe they could just take one for the team? Maybe we build a society focused on helping people instead of giving the rich yet another leg up they don't need?
...if you come home from working your second job and are ready to collapse, are you going to prepare a nutritious meal for your child and set them up with insightful, interesting media? No you're going to heat up chicken nuggets and put them in front of the iPad.
This is what I mean by "complicated factors of culture and habit." An iPad costs more than an assortment of paper books. Frozen chicken nuggets cost more than basic ingredients. But the iPad and nuggets are faster and more convenient. The kids-get-iPad-and-nuggets habit is popular with middle-income American families too, not just poor families where parents work two jobs. The economic explanation is too reductive.
I'm not trying to say that this is the "fault" of parents or of anyone in particular. When the iPad came out I doubt that Apple engineers or executives thought "now parents can spend less time engaging with children" or that parents thought of it as "a way to keep the kids quiet while I browse Pinterest" but here we are.
I was there (Apple) at the time. Absolutely did NOT expect this thing that Steve thought was a neat way to see the whole NYT front page at once, was going to be the defining MacGuffin of an entire generation of children.
I mean, that's the thing though. We now have had kids parked in front of iPads for a good amount of time, along with other technical innovations like social media, and we have documented scientific proof of the harms they do to children's self-esteem, focus and mental acuity. I don't think the designers of the iPad or even the engineers at Facebook set out to cause these issues, but. they. did. And now we have a fresh technology in the form of AI that whole swathes of "entrepreneurs" are ready to toss into more children's brains as these previous ones were.
Is it too much to ask for a hint of caution with regard to our most vulnerable populations brains?
As a former iPad (OS) designer, and former Facebook feed engineer, of course we're upset about what happened. Most of us fought valiantly, with awareness, against what became the dark forces and antisocial antipatterns. But the promo-culture performance incentive system instituted by HR being based on growth metrics at all costs made all of us powerless to stop it. Do something good for the world, miss your promo or get fired.
Circa 2020 a huge number of fed-up good-intentioned engineers and designers quit. It had no effect, at all.
I'm genuinely sorry that happened to you. That had to be an absolute nightmare of an experience.
To be clear: I am not saying that engineers need to be better at preventing this stuff. I am saying regulators need to demand that companies be careful, and study how this stuff is going to affect people, not just yeet it into the culture and see what happens.
But I have faith that people will notice the difference. The current generation may not care about autotune, but that doesn't mean another generation won't. People rediscover differences and decide what matters to them.
When superhero movies were new, almost everyone loved them. I was entranced. After being saturated with them... the audience dropped off. We started being dissatisfied with witty one-liners and meaningless action. Can you still sell a super-hero movie? Sure. Like all action movies, they internationalize well. But the domestic audiences are declining. It makes me think of Westerns. At one time, they were a hollywood staple. Now, not so much. Yes, they still make them, and a good one will do fine, but a mediocre one... maybe not.
> The current generation may not care about autotune
The previous generation's care about autotune was also flatly wrong. Autotune was used by a few prominent artists then and is more widely used now as an aesthetic choice, for the sound it creates which is distinctly not natural singing, as the effect was performed by running the autotune plugin at a much, much higher setting than was expected in regular use.
Tone correction occurs in basically every song production now, and you never hear it. Hell, newer tech can perform tone correction on the fly for live performances, and the actual singing being done on the stage can be swapped out on the fly with pre-recorded singing to let the performer rest, or even just lipsync the entire thing but still allow the performer to jump in when they want to and ad-lib or tweak delivery of certain parts of songs.
The autotune controversey was just wrong from end to end. When audio engineers don't want you to hear them correcting vocals, you don't hear it. I'd be willing to buy another engineer being able to hear tone correction in music, but if a layman says they do, sorry but I assume that person's full of shit.
There's already conversation in AI art about how "Y'all will miss all these weird AI glitches when they're gone!" It will become the new tape hiss. Something people will nostalgically simulate in later media that doesn't have it naturally.
But what they're describing is a case where someone with the storytelling ability but not the money or technical skills could create something that looks solid.
You're imagining "pls write film" but the case of being able to film something and then adjust and tweak it, easily change backdrops etc could lead to much higher polish on creations from smaller producers.
Would the green mile be any less hard hitting if the lights flickering were caused by an AI alteration to a scene? If the mouse was created purely by a machine?
I don't have a problem with adjusting small elements of the film, but that isn't going to make it a tool for youtubers (or home users off the grid) to tell their own stories.
You're unlikely to get an AI that wins accolades for the same reason that's unlikely with humans: they represent the absolute pinnacle of achievement.
The same AI can still raise the minimum bar for quality. Or replace YouTubers and similar while they're still learning how to be good in the first place.
No idea where we are in this whole process yet, but it's a continuum not a boolean.
What accolades? The Hollywood self-congratulatory conspicuous consumption festivals they use to show how good they are at producing "art" every year? The film festivals where billions of dollars are spent on clothing and jewelry to show off the "class" of everyone attending, which people like Weinstein used to pick victims, and everyone else uses as conspicuous consumption and "marketing" media?
Pinnacle is not the word I'd use. Race to the bottom, least possible effort, plausibly deniable quality, gross exploitation, capitalist bottom line - those are all things I'd use to describe current "art" awards like Grammy, Oscars, Cannes, etc.
The media industry is run by exploiting artists for licensing rights. The middle men and publishers add absolutely nothing to the mix. Google or Spotify or platforms arguably add value by surfacing, searching, categorizing, and so on, but not anywhere near the level of revenue capture they rationalize as their due.
When anyone and everyone can produce a film series or set of stories or song or artistic image that matches their inner artistic vision, and they're given the tools to do so without restriction or being beholden to anyone, then we're going to see high quality art and media that couldn't possibly be made in the grotesquely commercial environment we have now. These tools are as raw and rough and bad performing as they ever will be, and are only going to get better.
Shared universes of prompts and storylines and media styles and things that bring generative art and storytelling together to allow coherent social sharing and interactive media will be a thing. Kids in 10 years will be able to click and create their own cartoons and stories. Parents will be able to engage by establishing cultural parameters and maybe sneak in educational, ethical, and moral content designed around what they think is important. Artists are going to be able to produce every form of digital media and tune and tweak their vision using sophisticated tools and processes, and they're not going to be limited by budgets, politics, studio constraints, State Department limitations, wink/nod geopolitical agreements with nation states, and so on.
Art's going to get weird, and censorship will be nigh on impossible. People will create a lot of garbage, a lot of spam, low effort gifs and video memes, but more artists will be empowered than ever before, and I'm here for it.
Any accolades, be that professional groups, people's awards, rotten tomatoes or IMDB ratings.
> Race to the bottom, least possible effort, plausibly deniable quality, gross exploitation, capitalist bottom line - those are all things I'd use to describe current "art" awards like Grammy, Oscars, Cannes, etc.
I find them ridiculous in many ways, but no, one thing they're definitely not is a race to the bottom.
If you want to see what a race to the bottom looks like, The Room has a reputation for being generally terrible, "bad movie nights" are a thing, and Mystery Science Theater 3000's schtick is to poke fun at bad movies.
> The media industry is run by exploiting artists for licensing rights.
Yes
> The middle men and publishers add absolutely nothing to the mix. Google or Spotify or platforms arguably add value by surfacing, searching, categorizing, and so on, but not anywhere near the level of revenue capture they rationalize as their due.
I disagree. I think that every tech since a medium became subject to mass reproduction (different for video and audio, as early films were famously silent) has pushed things from a position close to egalitarianism towards a winner-takes-all. This includes Google: already-popular things become more popular, because Google knows you're more likely to engage with the more popular thing than the less popular thing. This dynamic also means that while anyone will be free to make their own personal vision (although most of us will have all the artistic talent of an inexperienced Tommy Wiseau), almost everyone will still only watch a handful of them.
> Art's going to get weird, and censorship will be nigh on impossible.
Bad news there, I'm afraid. AI you can run on your personal device, is quite capable of being used by the state to drive censorship at the level of your screen or your headphones.
Extremely-heavily-CG movies already mostly look like shit compared to ones where they build sets and have props and location shoot, even if somewhat assisted by computer compositing and such (everything is, nowadays). [edit: I don’t even mean that the graphics look bad, but the creative and artistic choices tend to be poor]
The limitations of reality seem to have a positive effect on the overall process of film making, for whatever reason. I expect generative AI film will be at least as bad. Gonna be hard to get an entire well-crafted film out of them.
I'd love to see what'd happen if someone dumped the entire text of the Silmarillion or the Hobbit into one of these models. Assuming context windows and output capacity become large enough.
Especially primed by all the lord of the rings movies, for example; I could see the studio taking all the archived footage, camera angles, all the extra data that was generated in the creation of the films and feeding that into something like this model to create all kinds of interesting additional material.
From this and other comments, I get the impression that most on HN assume the tool will be used exclusively by people without any sort of artistic talent, either plagiarizing existing works and/or producing absolute dreck.
However, I see an interesting middle ground appear: a talented writer could utilize the AI tooling to produce a movie based upon their own works without having to involve Hollywood, both giving more writers a chance to put their works in front of an audience as well as ensuring what's produced more closely matches their material (for better or worse).
> I wouldn't be. How is any of this going to lead to meaningful art?
Local art, local actors, local animations telling stories about local culture. A netflix for every city, even neighborhoods. That's going to be crazy fun.
There are plenty of great outsider storytellers and artists. Youtube is proof of this. People mostly do comedy on youtube because that's what the medium supports on a low budget, but AI is going to change that.
I'm really not seeing how that would happen from these examples. It would seem like achieving an adequate, directorial level of control would require writing a novel -- or, anyway, more than a conventional screenplay -- to get the AI to make the movie you wanted.
There is so much that has to be conveyed in making a film, if you want it to say something particular.
> It would seem like achieving an adequate, directorial level of control would require writing a novel -- or, anyway, more than a conventional screenplay -- to get the AI to make the movie you wanted.
And? What's the problem with that? You seem to be locked in a "prompt to get a movie" mindset.
Those are the examples provided. When they deliver pro tools for generating movie clips, I will be more convinced, but that hasn't remotely happened yet.
Looking back on history I think this will lead to meaningful art (and tons and tons of absolute garbage!).
The printing press led to publishing works being reachable by more people so we got tons of garbage but we also got those few individual geniuses that previously wouldn't have been able to get their works out.
I see similarities in indie video/PC games recently too. Once the tech got to the point that an individual or small group could create a game, we got tons of absolute garbage but also games like Cave Story and Stardew Valley (both single creators IIRC).
Anything that pushes the bar down on the money and effort needed to make something will result in way more of it being made. It also hopefully makes it possible for those rare geniuses to give us their output without the dilution of having to go through bigger groups first.
I'm also excited from the perspective that this decouples skills in the creative process. There have to be people out there with tremendous story telling and movie making skills who don't have the resources/connections to produce what they're capable of.
The printing press enabled the artistic visions of single individuals (the writers) to find a wide audience.
To do something similar, this has to allow the director (or whomever is prompting the AI) to control all meaningful choices so that they get more or less the movie the intend. That seems far away from what is demonstrated.
> How is any of this going to lead to meaningful art?
It's a powerful tool. A painting isn't better because the artist made their own paint. A movie made with IRL camera may not be better than one made with an AI camera.
We already exist in a world where most of the revenue for film companies comes from formulaic productions. Studio execs certainly worry about how they are going to create profits in addition to any concerns about the qualitative cultural value of the films.
If it becomes easy to make "Deadpool & Wolverine", it will no longer be where the money is. Everything that becomes a commodity attracts competition and ceases to be special. (You can see some of that in super-hero movies, which have started to be generic and lost some of their audience.)
But, in reality, even making that kind of film is miles away from these examples.
> If it becomes easy to make "Deadpool & Wolverine", it will no longer be where the money is. Everything that becomes a commodity attracts competition and ceases to be special. (You can see some of that in super-hero movies, which have started to be generic and lost some of their audience.)
Well, given the studios still hold the copyright, they can severely constrain supply to keep profits up.
My suspicion is that this kind of stuff gradually reduces some of the labor involved in making films and allows studios to continue padding their margins.
Counterpoint: home "studio" recording has been feasible for decades, but music execs are not ruffled. Sure, you get a Billie Eilish debut album once a generation, but the other 99.99% of charting music is from the old guard. The media/entertainment machine is so much bigger than just creating raw material.
Does anyone other than PMs when thinking up user stories do shit like this or finds this kinda stuff desirable? It just sounds like a business person who doesn't have a life other than selling their product trying to think up "real user" usecases every time.
Many creative works these days require the effort and input of so many people, so much time, and so much money that they can't have a specific creative vision. Mediums like book, comics, indie movies, and very low budget indie games, where the the end product was created by the smallest number of people, have the most potential to be interesting and creative. They can take risks. This doesn't mean they will be good, most aren't, but it means that the range of quality is much broader, with some having a chance to shine in ways which big budget projects just can't. The issue with small teams and small budgets is that they are inherently limited in what they can create. Better tools allows smaller groups of people to make things that previously would have required an entire studio but without diluting the creative vision.
Will this also result in a tidal wave of low effort garbage? Of course it will. But that can be ignored.
Nope. This is just like when cryptobros would regularly insist that cryptocurrencies would replace banks by the end of the decade. It's safe to assume that anyone who makes such wild predictions is a bagholder who stands to gain financially from said wild predictions coming true, even though they never will.
This sounds gimmicky and worth watching once or twice, then forgetting about. Worthwhile art will continue to be created from a specific person's/group's vision, not an algorithmically generated sum of personal preferences.
I rarely watch movies or read books twice anymore. There's too much content already. The challenge with purely human art at this point is that it will be silenced by the perpetual flood of half-assed generated work. There will be room in elite art circles for more, but at some point the generated stuff will be so ubiquitous (and even meaningful) that anyone without connections is going to have a tough time building an audience for their handcrafted work, unless it happens to be particularly controversial or 'difficult' to make. The demand for visual stimulus will be satisfied by hypertuned AI models. Generative AI is not there quite yet but there's no reason to think it won't be better than 90%+ of purely human content within a decade given the pace of development over the last few years.
I don't buy this narrative at all. People like people and increasingly follow artists because of their personality and overall "brand." No one cares about generated AI art or its creator(s), because it's not interesting. It's also not sharable with other humans; see, for example, the frenzy around going to a Taylor Swift concert. The mass appeal and shared interest is part of the draw.
At best, you'll get something like a generic sitcom. The idea that "all visual stimulus will be satisfied by hypertuned AI models" doesn't line up with how people experience the arts, at all.
That may be the case today but kids are starting to grow up with this stuff as part of their lives, and I don't think we can anticipate the reaction as both they and the models grow in tandem. I think human creativity is much deeper than LLMs, but that is from my human perspective and I can't fully rule out that the LLMs may become better at it at some point in the future. I actually think they're already smarter and more creative than most people (though not more than the potential of any given human if they practiced/trained thoroughly).
I fully agree here. I want to be part of an audience, and as part of that audience I always look at the human development of the things to share - artifacts in the case of fine art, or experiences in the case of performative art. The artist will always be more important than their work to me.
I don't want to carry mechanical solutions labelled culture - deterministic enough, despite hallucinations - into the next generation that follows my own. It's an impressive advancement for automation, sure, but just not a value worth sharing as human development.
That being said, I think GenAI could be a valuable addition in any blueprint-/prototype-/wireframing phase. But, ironically, it positions itself in stark contrast to what I would consider my standards to contemporary brainstorming, considering the current Zeitgeist:
- truthful to history and research (GenAI is marketing and propaganda)
- aware of resources (GenAI is wasteful computing)
- materialistic beyond mere capitalistic gains (GenAI produces short-lived digital data output and isn't really worth anything)
I think that’s an extremely short sited perspective. There isn’t much that separates a “fun and wacky” movie from something impactful from a cinematography perspective. With the right music, ambience and script you could absolutely do any genre of movie you wish to.
I disagree, I believe your perspective is short-sighted. If you really think what makes a movie "impactful" is the music, ambience, and script then I don't think you have much media literacy.
It's no more ridiculous than saying what makes a painting impactful is the brush strokes. But if I copy Picasso's work stroke for stroke, why am I not Picasso? After all, the dumbass paints like a child, admittedly! How could someone like him ever be considered a great painter?
If they are stitching then I would consider that a form of art.
However, merely describing something is not doing the thing. Otherwise, the business analysists at my company would be software engineers. No, I make the software, and they describe it.
The end-goal here is humanless automation, no? Then I'm not sure your assumption holds up. If there's no human, I question the value.
You may question the value but if it’s anything like rugs you won’t be in the majority. People pay a significant premium for artisanal handmade rugs but that being said, more than 95% of the rugs people use are machine made because they’re essentially indistinguishable from a handmade one and are much, much cheaper and just as functional.
I'm sure eventually you can, but I don't think triggering emotions is the correct "KPI" so to speak.
On social media platforms, typically the most popular content triggers the strongest emotions. It's rage-bait however, or sadness bait, or any other kind of emotional manipulation. It tricks the human mind and drives up engagement, but I don't think that is indicative of its value.
To be clear, I'm certain that's not what you're doing, and the music is good. But I think it's complicated enough that triggering emotions isn't enough data to ascertain value.
I don't know, exactly, what combination of measurements are needed to ascertain value. But I'm confident human-ness is part of the equation. I think if people are even aware of the fact a human didn't make something they lose interest. That makes the future of AI in entertainment dicey, and I think that's what fuels the constant dishonesty around AI we're seeing right now. Art is funny in how it works because, I think, intention does matter. And knowledge about the intention matters, too. It maybe doesn't make much sense, but that's how I see it.
Right now there is a ton of stigma around AI art. That stigma fuels a ton of poorly-informed rhetoric against it. There is also tons and tons of casual use of AI art being shitposted for funsies everywhere that reinforces that rhetoric that AI art means "Push button. Receive crap. Repeat."
Meanwhile, as someone who has been engaged with the AI art community for years, and spent years volunteering part-time as a content moderator for Midjourney, the process of creating art via AI with intentionality is deeply human.
As an MJ mod, I have seeeeeeen things.... It's like browsing though people's psyche. Even in public portfolios people bare their souls because they assume no one will bother to look. People use AI to process the world, their lives, their desires, their trauma. So much of it is straight-up self-directed art therapy. Pages and pages, thousands of images stretching over weeks, sometimes months, of digging into the depths of their selves.
Now go through that process to make something you intend to speak publicly from the depth of your own soul. You don't see much of that day to day because it is difficult. It's risky at a deeply personal level to expose yourself like that.
But, be honest: How much deeply personal art do you see day to day? You see tons of ads and memes. But, to find "real art" you have to explicitly dig for it. Shitposting AI images is as fun and easy as shitposting images from meme generators. So, no surprise you see floods of shitposts everywhere. But, when was the last time you explicitly searched out meaningful AI art?
> But, be honest: How much deeply personal art do you see day to day?
You bring up a good point - very little. But, to be fair, those people aren't necessarily trying to convince me it's art.
I think you're mostly right but I am a little caught up on the details. I think it's mostly a thing of where the process is so different, and involves no physical strokes or manipulation, that I doubt it. And maybe that's incorrect.
However, I will also see a lot of people who don't know how to do art pretending like they've figured it all out. I also see the problem with that. It wouldn't be such a problem if people didn't take such an overly-confident stance in their abilities. I mean, it's a little offensive for that guy mucking around for an hour to act like he's DiVinci. And maybe he's a minority, I wouldn't know, I don't have that kind of visibility into the space.
I think a lot of the friction comes from that. Shitposts are shitposts, but I mean... we call them shitposts, you know? They, the people that make them, call them shitposts. There's a level of humility there I haven't necessarily seen with "AI Bros".
I think, if you really love art, AI can be a means to create a product but it can also be a starting point to explore the space. Explore styles, explore technique, explore the history. And I think that might be missing in some cases.
For a personal example, I'm really into fashion and style. I love clothes and always have. But it's really been an inspiration to me to create clothes, to sew. I've done hand sewing, many machine stitches too. And I don't need to - I could explore this in a more "high-level" context, and just curate clothing. But I think there's value in learning the smaller actions, including the obsolete ones.
"Turn this book into a two-hour movie with the likeness of [your favorite actor/actress] in the lead role."
This sounds marginally above fanfiction, so I do think it'll be very easy to tell them apart. "Terminator, except with Adam Sandler and set on Mars" is a cute, gimmicky idea, not a competitor to serious work.
Well, yeah. If you explicitly try to come up with a cute, gimmicky idea, it's not going to be serious. Taste still matters regardless of paint, cameras or computers.
Maybe, but I guarantee you this is going to get banned in the US for "safety" or "misinformation" reasons eventually (with large backing from Hollywood).
Not really. A vision implies a particular kind of project, presumably created by someone with expertise and some well-thought through ideas about what it ought to be. Personal preferences just mean that someone likes X qualities.
To use a real-world example: if the Renaissance-era patrons had merely written down their preferences and had work made to match those preferences, it's highly unlikely that you'd have gotten the Mona Lisa or David.
Which is to say that, there will definitely be some interesting and compelling art made with AI tools. But it will be made by a specific person with an artistic vision in mind, and not merely an algorithm checking boxes.
In enthusiasm perhaps, but when I play that in my own head with the voice of Dr. Károly Zsolnai-Fehér, he doesn't script his videos like that. I can't recall a single instance of him using triple repetition like that.
I think AI porn is overhyped. We've had the ability to create realistic photos and short vids for over a year now and onlyfans creators are still doing fine. AI porn is just a niche for stuff that humans don't want to perform.
I think AI porn is chocked full of fascinating moral quandaries. It kind of transcends all other types of GenAI for the amount of hard questions it asks society.
As they say, porn is always the leading spear of technology. It's something to keep an eye on (no pun intended) to understand how society will accept/integrate generated content.
It's definitely interesting from a moral and tech perspective.
However, commercially it seems like a niche within the existing structures of porn. Mostly competing with the market for animated stuff. At least that's where it is right now, and its already at photorealistic parity with human content creators.
We have not had the ability to create interesting ai porn vids yet. How would we? Meta just showed movie gen.
But i'm pretty sure the short images of moving woman very subtle might gotten the one or other of. Just wait a little bit until you can really create wat you are looking for.
You obviously have not spent any time on civit.ai if you're saying that.
What scares me most is that in my opinion, by far the best prompt writers are the ones who are deeply "motivated" and "experienced" with prompting. Often the best prompters have only one hand on the keyboard at a time.
You have no idea how much butthurt their is from specifically artists who draw at the AI NSFW models which exist.
I can trivially fine-tune and create more art from certain artists in an hour than they have produced themselves in their whole careers. This makes a lot of people very upset.
> Sooner than anyone could have expected, we'll be able to ask the machines: "Turn this book into a two-hour movie with the likeness of [your favorite actor/actress] in the lead role."
I've been doing this with ChatGPT, except it's more of a "turn into a screenplay" then "create a graphic of each scene" and telling it how I want each character to look. It's works pretty well but results in more of a graphic novel than a movie. I'm definitely been waiting for the video version to be available!
It's only going to get better, faster, cheaper, easier.[a]
Sooner than anyone could have expected, we'll be able to ask the machines: "Turn this book into a two-hour movie with the likeness of [your favorite actor/actress] in the lead role."
Sooner than anyone could have expected, we'll be able to have immersive VR experiences that are crafted to each person.
Sooner than anyone could have expected, we won't be able to identify deepfakes anymore.
We sure live in interesting times!
---
[a] With apologies to Daft Punk: https://www.youtube.com/watch?v=gAjR4_CbPpQ