The most ambitious AI film I've made to date.
We are less than a year out from commercially viable AI films. Updates to
@runwayml
(control) and
@elevenlabsio
(speech-to-speech) have me believing it.
If you liked it, some of the people that inspired me recently are…
How are things moving THIS FAST!?
There's been a major breakthrough in AI animation, announced as part of the Animate Anyone project.
They seem to have solved the major image-to-video challenges with a new framework I'll share more on in the comments.
#aianimation
#aivideo
…
Critics are calling The Lost Father "a piece of shit slide show", "pick up a pencil", and "horrendous AI dogshit".
Dive into a world of poignant AI dogshit with our latest animated film. Experience the journey of a lifetime, premiering exclusively on Twitter. Don't miss out on…
This new AI workflow kind of blew my mind.
A Redditor named lotrfan2004 recreated a Pulp Fiction scene as a Playstation 2 game. Here's how:
1) They used
@aimirror_app
to style transfer stills of the characters
2) They uploaded the stylized stills of the characters + Pulp…
@vproceart
Is this a real story?
Removing people from that image would be one promptless roll of generative fill in Photoshop.
You’re telling me a studio hired and on-boarded image prompters without the ability to use the easiest part of photoshop? Then fired them immediately?
I’m not…
Runway's seemingly approaching AI video from a different angle than Pika: director control.
This is a preview of the increased camera control you'll soon have. They're enabling real filmmaking.
#RunwayGen2
#runway
#aivideo
Well, this is awesome. You can now do long-form, 3D AI videos like this without Stable Diffusion, or more than a few minutes work.
This is only my first stab at it.
Prepping a YouTube tutorial on it now.
#aicinema
#aiartcommunity
#aivideo
I'm adjusting to the new Pika 1.0 interface vs. what I'm used to in Discord.
New UI is absolutely slick. They were right to sit tight on this a bit until it was polished.
#PIKALABS
@CorralSummer
@runwayml
@elevenlabsio
Just for the sake of being more clear on where the law is at to readers:
Elements of works created by artificial intelligence (AI) are not protected by copyright. However, if a human edits or manipulates the AI-generated work in a "sufficiently creative" manner, the resulting…
When
@runwayml
ships a new feature, it's always a leap forward in generative AI video.
I've been playing around with early access to the new Multi Motion Brush. 🔊🔛
What a f**king upgrade in letting you control movement!
#aicinema
#aiartcommunity
#runwayml
As this crosses 4m views, I want to be cheeky about the free publicity, but it’s clear real artists are feeling very affected by AI.
Most of us AI tinkerers are generally supportive of your concerns, particularly around copyright and compensation. Beyond fan art/parody, this…
Wanted this to feel as much like a real, campy thriller — as far as the AI tools let you get.
Dialogue is still hard, but it's slowly getting more realistic.
#aivideo
#aimovie
I don't work for Runway, but have to say
@c_valenzuelab
is an inspiring CEO.
Whether or not you think Runway can compete with OpenAI in the short timeframe they're gonna have, the signal he's giving to employees and the market is that Sora is exciting for him — not threatening.
One thing I didn't expect with Midjourney's new Character Reference feature:
It's not limited to one character at a time. 🧑🤝🧑
Not really sure how I ended up with a classic Disney Italian soap opera, but here we are.
#midjourneyArt
#midjourney
#aiartcommunity
It's my first time trying out
@everartai
to create a consistent character.
What I fed the model wasn't even ideal — a quick test run.
This is going to be a game-changer for filmmakers.
#aianimation
#AIanime
#AIart
Because
@Runway
released incredible lip-sync today it’s time I confess something:
I’ve been working on the greatest protest song of our generation. Made possible by
@suno_ai_
I can’t share the whole thing until a big project I’m doing related to Sleepy Hollow comes out —…
I'm really excited to say I'm now part of the Leonardo Creator Program!
@LeonardoAi_
's product velocity has been incredible, and recently I've been commissioned to work on some big AI anime projects that their tools are going to help a lot with.
Very pumped to dig into their…
This clip is from one of my favourite animated movies of all time.
To get the lighting right, it took many grueling days...oh wait, wrong clip. This was just done in Runway Gen 2
#aianimation
#aivideo
#RunwayGen2
Wanted this to feel as much like a real, campy thriller — as far as the AI tools let you get.
Dialogue is still hard, but it's slowly getting more realistic.
#aivideo
#aimovie
The most ambitious AI film I've made to date.
We are less than a year out from commercially viable AI films. Updates to
@runwayml
(control) and
@elevenlabsio
(speech-to-speech) have me believing it.
If you liked it, some of the people that inspired me recently are…
They said it was too controversial, too groundbreaking.
But the disgusting acts of pesticide in the flower kingdom need to be exposed.
Presenting: 👑 THE FLOWER KING 🌸
Witness the shocking, untold saga of ambition, treachery, and floral conquest. Made with
@suno_ai_
…
The Terra Age is an official finalist for the
#runwayml
#gen48
competition!
It's dystopian Sci-Fi about two terraformed planets low on resources that need to start a war for survival — from a brain with 48 hours of no sleep.
Please vote for it via the link in the…
This might be the first time this has been posted in full on X.
I’m thirsty for Sora, but already feeling nostalgic about how cleverly the 1st gen AI filmmakers worked through harsh limitations.
This is the last few months of that era.
#aicinema
#aivideo
#AIArtistCommunity
Share your creations. DMs are open.
Title: Synthetic Life Description: When a virtual reality product eliminates loneliness, its widespread adoption reshapes society, trading utopian happiness for dystopian dependency.
Done by
@TheReelRobot
#aifilms
#aifilm
#aimovies
…
Well, this is awesome. You can now do long-form, 3D AI videos like this without Stable Diffusion, or more than a few minutes work.
This is only my first stab at it.
Prepping a YouTube tutorial on it now.
#aicinema
#aiartcommunity
#aivideo
I tested the
@runwayml
Gen 2 update:
textures, reflecting surfaces, underwater, 2-sided photos, animation, fish eye lens, humans + animals in same scene, etc.
The scientific conclusion is that it is awesome.
2nd and 3rd shots weren't possible before without morphing.…
Alright, think I'm cracking the code on dialogue in AI Videos with more movement and background animation.
@Uncanny_Harry
paved the way.
Very rough first try, but I'm going to work on it and share a tutorial when I've fleshed it out.
#PIKALABS
#HeyGen
#aivideo
An AI Filmmaking GPT to automate Midjourney work, 10x consistency, and more.
Tutorial + link up on YouTube.
This Christmas, give her the gift of time.
#aivideos
#aicinema
#midjourney
I started a new routine combining mental health + fresh powder for AI art:
1) With just my iPhone, I'll go on a walk and shoot some local nature photography (photo on left)
2) I take it to Midjourney, and use /Describe to see how MJ would prompt for it (check alt)
3) I drop it…
Tonight I took a photo of my wife by a pool with lanterns, then took it to Midjourney and Magnific AI as a reference image.
Nothing is as satisfying or original in output as using your own photos as reference images
#AIArtwork
#midjourney
#aiartcommunity
Used
@runwayml
's director mode to get a long take of a giant creature emerging. SOUND ON
Little details aren't perfect yet, but working on my animation course has made it clear that Runway's reallllly figuring out generative video.
#AiAnime
#aivideo
#aianimation
In a nutshell, they came up with a new framework. They take an image, figure out character movements for it, then introduces some advanced techniques (like adding digital 'noise' for detail and using a special process to clear it up) to create the video.
It focuses really…
Now that we can create incredible AI music with
@udiomusic
, can we please use it for the right reasons:
Bringing back peak 1990s
@MentosUSA
commercials?
[note, this is original Mentos lyrics — but song is made with Udio]
#mentos
#thefreshmaker
#AIMusic
One thing I didn't expect with Midjourney's new Character Reference feature:
It's not limited to one character at a time. 🧑🤝🧑
Not really sure how I ended up with a classic Disney Italian soap opera, but here we are.
#midjourneyArt
#midjourney
#aiartcommunity
@AIandDesign
@runwayml
Animation artists who’ve been treated like a cog in the wheel will be creating their own movies soon.
Change is less scary when people can see the benefits.
BREAKING: Runway is about to release a game-changing feature for Gen 2: Moving With Intention
This thread is a preview of what you can do with it:
#aivideo
#MidjourneyAI
#RunwayGen2
This new AI workflow is going to be good for anime in particular.
20 minutes (not including rendering time) is about how long it takes to get an action scene, with 1-2 characters, of up to 30 seconds.
Will post a longer clip later, and the YT tutorial in a couple of days…
Lately, I work on a lot of animated content for clients using AI.
Many argue AI’s hurting professional animators, but every stat I’ve seen shows they’re generally overworked to the point of misery.
We are arriving at a real solution to their problems, and somehow only AI geeks…
Apple's getting into AI animation.
It’s not going to be the same as Runway or Pika Labs — it’s more like an AI-powered After Effects or 2D animation tool.
The prototype they made is called Keyframer. They use OpenAI’s GPT-4, so you can prompt (e.g. “make the stars…
@mogustine
@runwayml
@elevenlabsio
Getting good lip sync with 2D images is hard. It’s why I didn’t do very much.
That, and consistency. It’s a lot of re-rolls, credits, time, and photoshop.
🎥🎞️🎧 저녁 식사 (The Dinner)
The Dinner, is a 5-character, AI-gen film mixing vid2vid & img2vid. All footage is shot in Ct UE5 with multiple cameras in a virtual 3D env & green screen & stylized with Kaiber. Lipsync is done with Runway, as are the img2vid seq. Voices are made…
@chaseleantj
Spend more time finding good reference images than on your prompts.
Your taste is the most important skill to develop with AI art.
Find images that speak to your taste, figure out how you’d do it different, and go low —iw. It’ll usually be more unique than a text only prompt.
@JeffSynthesized
Yep, this is probably the new best AI Video to date. 👏
So nice to see pacing that’s not limited by the shots we get from generative video.
The exterior shots of the spaceship were done in After Effects?
@c_valenzuelab
Yep. You guys have the expertise, singular focus, and a world of possible competitive differentiation.
Get them technical papers and ye shall have the crown.
Curious about all the hype around AI video? 🎥
I just dropped a master post on the space.
It covers where to generate video, how the models compare, technical challenges yet to be solved...and much more 👇
@GraceaBlout
@DUAJISIN
@CorralSummer
@runwayml
@elevenlabsio
Yes, you’ve shared a biased article, where the subject is a comic book made with Midjourney art. (There’s good unbiased articles on it too, btw).
Beyond the fact the ruling only partially canceled it, I agree that the more derivative it is, the less it should be copyrightable.…
@Skullgrin14022
I don't expect we're going to come to an agreement, but I have a different worldview on it.
I agree some harm is being done. I actually don't like, for example, that the post I shared had some obvious Studio Ghibli elements. Those images are clearly derivative and I hope the…
My AI Film won a film fest award.
It's no contender for Sundance, but I think we're seeing the start of a sea change where AI films can compete with traditional ones.
#aivideo
#runwayml
#aiartcommunity
@runwaymlstudios
A silohuette of a man holding a cell phone in the dark outside a farmhouse, shadows, night sky, amber glow of phone is the only light, frontlighting, desaturated greens, shallow depth of field, medium shot, 32k UHD, ARRI Alexa
@Jason
To be fair to them
@Jason
— we have no social media posts from Gen X at that age.
If we did, we could cherry pick some underdeveloped kids from any generation.
Happy to see Reddit was a fan of this AI-generated Star Wars film.
Runway and Pika Labs can do exceptional portraits, but action and advanced movement is where creators can get stuck. Getting moving AT-ATs took some patience.
#AIart
#GenerativeAI
#MidjourneyAI
@iamneubert
I’m recording a full lesson on it for my YT audience atm, but I basically spent some time breaking down the color palettes, lighting, and composition of Ghibli animations, and copy/paste the hell out of those tokens in my prompts.
Ends up being more effective than just writing…
I'm honored to be featured in Dale's new video. He dedicated more than 30 hours to this project, and it shows. The video is incredibly valuable, and I gained a lot of insights from it.
Full video in the thread 👇