Yea. But like. How do you act with it?
this post was submitted on 10 Oct 2025
43 points (93.9% liked)
movies
1885 readers
353 users here now
A community about movies and cinema.
Related communities:
- !television@piefed.social
- !homevideo@feddit.uk
- !mediareviews@lemmy.world
- !casualconversation@piefed.social
Rules
- Be civil
- No discrimination or prejudice of any kind
- Do not spam
- Stay on topic
- These rules will evolve as this community grows
No posts or comments will be removed without an explanation from mods.
founded 7 months ago
MODERATORS
By " prompt engineering " I guess
Did anyone ask for AI actors? That's a genuine question. It's def not something I'm interested in seeing. I want Ai mowing my lawn, doing my laundry, cleaning my house (already doing so with robot vacuums somewhat)....shit I don't want to do.
I fully endorse photorealistic cartoons. Characters can look like anything, without having to find a specific guy and hope fits the role.
Getting the cartoon onscreen can still involve an actual actor. Diffusion turns whatever you have into whatever you describe. Turning a guy into another guy is not a big ask. It's how some of this mess started, with Nicolas Cage deepfaked as Superman, and defictionalizing the Stallone version of The Terminator. The target face does not need to be a real person. Three actors can stage nearly any script.
Same goes for voice-acting. VAs are understandably concerned about being cloned. Nobody's talking about the opposite: making up what characters sounds like, so any actor can play anybody. Or everybody. You can even substitute, when a scene needs extra oomph - like a band featuring a guitarist for a solo. Same sound... distinct performance.
Same goes for voice-acting. VAs are understandably concerned about being cloned. Nobody’s talking about the opposite: making up what characters sounds like, so any actor can play anybody. Or everybody.
If anyone can play a role, there is no longer any need for voice actors.
Changing your voice won't fix bad acting.
The voice is the acting. It's in the name.
Like a character's just a face.
Everyone's got a Homer Simpson impression. Very few of them sound like Dan Castellaneta. This tech fixes how your vocal cords are shaped - not whether you can pull off an American accent.
You dont need AI to pitch shift a voice.
Uh huh. So it's more than naive pitch-shifting, but less than somehow fixing "oh god oh man oh man oh god." Like how someone sounds is more complex than playback speed, but still distinct from how they choose to say things.
You can figure this out. I believe in you.
That's called acting.
Voice acting is acting.
The name is a hint.
But acting skill alone won't let Idris Elba sound like Tilda Swinton. AI can.
Her creator, Eline Van der Velden, the CEO of AI production company Particle6, insists that Norwood was never meant to replace a real actor.
I keep seeing this quote, and I just don't understand how it's anything other than a blatant, outright, clear, and obvious lie.
If it wasn't meant to replace a real actor... Then what the fuck was it for?
I assume it will be doing porn in 5, 4, 3 ..
That's been happening. Source Film Maker (SFM) is a tool used to make movies from video game characters, but there are repositories that have characters from much more than just video games, and a lot of them have nude models. So they've been making porn with computer-generated characters for a while, no AI needed. AI will just make it easier, and the creation of such accessible to more people (e.g. take this model and make her do this that and the third).
The bright side (or the dark side, I suppose it depends who you ask) is that you can ask the AI or CGI character to do things you might not feel comfortable asking a human actress to do. Again, I'm not sure if that's a net good or bad for humanity, but at least it's just a computer doing it, right, and not a human woman with feelings?
The issue with the last point is that many people already have difficulties separating actors from roles. I can't unsee Astin's Samwise Gamgee and even when Blanchett played Hela in the MCU I had to stop thinking about her as Galadriel, because those are the characters burned in my mind for those faces. And I'm far from the worst there: We've had plenty of examples of people getting hated for a character they put on the screen, or stalked or otherwise pigeonholed into a single role.
I worry that this will only extend into the human-/GenAI-made distinction. People already make porn of existing persons, sexualise their feet and more. Increasingly advanced GenAI will make that more prolific. I worry that a generated facsimile of some real person will lead to the "original" getting associated with and harassed over whatever thing the prompter made the former do, particularly with things you might not feel comfortable asking a human actress to do.
The knock-on might be that human actors may increasingly feel like they have to also do things you might not normally feel comfortable asking them to do in order to compete. The "I prefer real people making my porn/movies" appeal may be an edge, but only for as long as "real people" also make the stuff I like to see.
Worse still, the actors might not personally feel that way, but Hollywood doesn't always run on free will, and I'm not optimistic that the adult movie industry is all consent and sunshine. If the agent thinks there's more money to be made...
The flipside of putting an actor's face on smut is that you could just as easily put any face on an actor. Cate Blanchett's characters won't have to look like one another any more than Grey DeLisle's characters look like one another. Cast your leads from This Person Does Not Exist. Like if Iron Man had Robert Downey Jr's acting, but looked like the Avengers video game version.
Given that porn has been at the forefront of technological innovation for at least the last fifty years, it's telling we haven't been deluged with AI-generated porn.
Uh.
Have you checked?
Civitai has entered the chat...
I take your point though, it's not gone mainstream as far as I know
Oh my god lol now that might stop it...
Or we will have found the main use case for AI
Whoa it just went straight crack cocaine! They must have trained it with The Parent Trap and Freaky Friday.
...from what we’ve seen, audiences aren’t interested in watching computer-generated content untethered from the human experience.
My sweet summer child have you never seen a toddler with an iPad?
This whole thing was one guy trying to put his cartoon OC into a labor union. It's a stunt that's wrong for reasons unrelated to anybody's hate-boner against all things AI.
The "digital actress" from Final Fantasy: The Spirits Within would be identically unqualified. Characters aren't persons.
That's just a cartoon character, though. The woman from the Final Fantasy movie, I mean. She was completely controlled by humans.
An AI actress would be able to be handed a script and could actually do the things without humans having to model her movements. The computer is doing the same work on the back end, but it takes a lot less human talent and creativity to do so.
Think about what it would take to get a computer-animated woman to walk down a street or up a hill. Now imagine something complicated — this is gonna sound weird, but the first thing that came to mind, when she was 10-11, my sister had this trick she'd do, our mom would toss her a shirt and tell her to try it on. Rather than go behind a closed door and change like a normal person, but also without exposing her chest to my brother and I, or asking us to turn around, she would put the new shirt on on top of the old one, and do a little dance and toss the old shirt aside, exposing nothing (maybe some tummy). A lot of girls could do it. I think it involves pulling the arms in both sleeves and slipping the inner one out and over the neck, but I never tried it. It's probably way simpler than it looks. Nevermind that though. It's way weirder than it sounds. It's something preteen girls know how to do, though. So imagine having to animate that. As opposed to having an AI actress just do it. So you have to tell the AI how the human body moves. Elbows, fingers, aren't double jointed etc., you have to have limits on the range of motion. Can't cheat, can't touch your elbow with the hand on the same side. Also, even under the top shirt, the bottom shirt has to obey the law of physics. It can't clip through the arms or the breasts, and the whole thing has to fit within certain laws of physics. Still, given all that, the AI should be able to just do it. A real person might have to get on a webcam and show the AI the maneuver, but given a couple examples (which it may even be able to find on the Web), it can do it. Then you just have the scene written like, her friend gives her a shirt to try on, so she puts it on, slips out of the old one without exposing any private areas, and twirls around, showing it off — and the AI just does it.
I would be shocked if any diffusion model could do that based on a description. Most can't overfill a wine glass.
Rendering over someone demonstrating the movement, as video-to-video, is obviously easier than firing up Blender. But: that's distant from any dream of treating the program like an actress. Each model's understanding is shallow and opinionated. You cannot rely on text instructions.
The practical magic from video models, for the immediate future, is that your video input can be real half-assed. Two stand-ins can play a whole cast, one interaction at a time. Or a blurry pre-vis in Blender can go straight to a finished shot. At no point will current technologies be more than loose control of a cartoon character, because to these models, everything is a cartoon character. It doesn't know the difference between an actor and a render. It just knows shinier examples with pinchier proportions move faster.
I really don't know what we can do with AI today. I do know that what we can do today seemed a distant dream not too long ago. It's moving fast and I can't imagine how far along it'll be in one year, or even five.
... you should probably check, before you go selling the what-ifs.
Diffusion is a denoising algorithm. It's just powerful enough that "noise" can mean, all the parts that don't look like Shrek eating ramen. Show it a blank page and it'll squint until it sees that. It's pretty good at finding Shrek. It's so-so at finding "eating." You're better-off starting from rough approximation, like video of a guy eating ramen. And it probably doesn't hurt if he's painted green.
There are a lot of extra hands on that railing. Also her eyes are lopsided
gross