Runway Gen-3 STUNS Everyone! BEST AI Video Going Public | First Look

[@aisamsonreal]

Here is an in-depth review of Gen-3 and comparison with Luma kzbin.info/www/bejne/h2S3e4eshdyij7s

[@tiffanyvelky]

Spectacular narration! Good job Human:)

[@AIverse111]

This is INSANE! cant wait o get my hands on this! BTW congrats to 100K subscribers!

[@aisamsonreal]

Me too! And thanks, I appreciate that!

[@musigxYT]

Gen-4 will definitely be the time when most people get fooled by videos. Great time to be alive.

[@aisamsonreal]

It really is!

[@spookyacti]

"dear fellow scholars" is gonna love this

[@YouTubeDeletesMyCommentssss]

Do you not understand the repercussions of people not knowing what to believe? the internet will crumble, be harshly regulated because of so much false information being out there.

[@himanilsharma2147]

You are typing this comment on wrong channel 😂

[@tbuermorei]

honestly.. idek lol

[@robertgoldbornatyout]

💯💯💯💯💯👍👍👍100k! Congrats! Thanks for keeping us up to date with the latest in the AI world.👍👍👍💯💯💯💯💯

[@bcarse1]

Great video Samson! Bravo. Loved the comparison at the end. Particularly admired your comments at the very end re: the AI revolution. I agree and am on Team Samson!

[@aisamsonreal]

Well said! Thanks for your support in this revolution!

[@RichardGolD-wz3is]

Not samson but Son of Sam 😂😂

[@blueberrywxx]

Thanks for your honesty and candidness.

[@aisamsonreal]

My pleasure!

[@I-Dophler]

The potential of Runway Gen 3 for generating immersive VR experiences is truly remarkable. With its ability to create consistent, high-fidelity environments and characters, we can envision a future where VR simulations are generated in real-time, transporting us into breathtakingly realistic virtual worlds. The possibilities for storytelling, exploration, and interaction within these simulated universes are boundless, offering an unprecedented level of immersion that could revolutionize the way we experience and engage with digital content.

[@aisamsonreal]

Totally agree

[@I-Dophler]

@@aisamsonreal Yes, very much.

[@littledovecitydust]

While the potential of Runway Gen 3 for creating immersive VR experiences is undoubtedly exciting, I have reservations about the implications for human creativity and uniqueness. Sure, AI can produce consistent, high-fidelity environments and characters, but can it truly replicate the emotional depth and nuanced storytelling that comes from a human mind? Every creator's perspective and experiences shape their work, resulting in diverse and rich content. If we rely too heavily on AI, we risk losing this diversity, leading to a more homogenized and less innovative digital landscape. The human touch in storytelling is irreplaceable, and it's essential to preserve that in our digital future.

[@CarlitoeGames]

ok miss GPT

[@I-Dophler]

@@CarlitoeGames OK, you must be human, madam xxxx.

[@carljhirst]

Soon be watching things like a Game of Thrones, every week, not knowing what might happen, or where the story is going. It just keeps going, script, art style and graphics and characters. It would be crazy to watch! If AI can do this now! What are we inside right now? We could be Uploaded Intelligences, in a AI Generated Reality. Not to Mention, the stories on the news could be fake. This is a new Age of Ai. Ai Revolution!!!

[@goldnarms435]

Dude you nailed it. In the not-too-distant future, someone like me will produce Marvel movies at home.

[@kayes4495]

AI can already do that, those in power are at least a decade ahead of the rest of us, it'll slowly be rolled out

[@carljhirst]

@@kayes4495 40-50 YEARS AHEAD.

[@sanjesco]

Great delivery.. good video. Thanks!

[@chariots8x230]

I hope we’ll have some camera controls when using this tool. I’d like to have control over the camera angle & camera movement.

[@eyadarabi4262]

Nice update! Still waiting to see artifacts-free and yet production quality level!

[@maxgwyn101]

100k! Congrats! Thanks for keeping us up to date with the latest in the AI world.

[@aisamsonreal]

Thanks, I appreciate it :)

[@chariots8x230]

Combine something like this with image to video, lip-syncing, consistent characters, and ability to have multiple consistent characters interacting with each other in a scene. That’s when we’ll be able to create full films with AI.

[@aaronjaggars321]

Hire me I can lil busy rn tho

[@richoffks]

Films are way past dead. The future is us living out the movie for an undefined amount of time. We’ll be the character, and we’ll interact with the world to progress through the 100% unique story because everything will evolve based around our decisions

[@misob]

like we have audio books from regular books, i think soon we can maybe have video books as well 😀

[@punchtalestudio]

Chris Marker made La jetée in 1969, a cult b&w photo movie that Terry Guilliam adapted in 97 with 12 monkeys. Stop thinking in old format, experiment new narratives. Unless u have nothing to say in particular

[@Antares-vj7su]

keep dreaming we are still very far from that

[@aisamsonreal]

Thanks for watching! What do think of Runway Gen-3? And what will you create?

[@elfe-ul2th]

have several screenplays i wrote over the years. Most of them were never produced since we couldnt fund them. Now we will be able to produce our movies in the near future without the huge budgeds that are needed. What a magnicifant tool to come! thanx for showing.

[@NewMarshallPlan]

This is so great! I want to use this to create music videos for my band where we feature ourselves in!

[@TheDandonian]

@@elfe-ul2th When open source reaches this level, the entire game changes.

[@Lateralusaint.]

@@NewMarshallPlan Great idea I formed a band with robots. Already making AI VIDEOS ;)

[@cuttothechasenews]

back in the day when movies first started. All they would be is a clip of a train going by, a horse running, all very menial things that would amaze people. Realizing the future those simple moving pictures would have, this stuff here reminds me of all of that.

[@Retro_Movie_Rewind]

I just created my first AI "zombie" film using Gen-3 and I have to say I had a blast. It was frustrating at first but once I got used to prompting like a movie director, the results quickly improved. Too bad it's so expensive. I don't know if I can afford to make another one.

[@lauriemcelroy8998]

Look forward to your runway G3 testing . I’m using 2 but have been deferring to Pika . I’ll get back on board with runway with this potential upgrade.

[@bigmikessenioradventures]

Great video keep up the good work. Subscribed!

[@mycloudvip]

Glad to be here!

[@carlhopkinson]

Unbelievably good.

[@I-Dophler]

Runway Gen-3 is a game-changing AI video generation model that leaves OpenAI's Sora in the dust. With its breathtaking cinematic visuals, seamless depiction of complex movements like running, and lifelike rendering of hands, Gen-3 has raised the bar to an unprecedented level. Its remarkable consistency in maintaining character appearances throughout videos is a true marvel. While not flawless, with occasional text incoherence and minor physics inaccuracies, Gen-3 has set a new benchmark, leaving open-source models like Luma Dream Machine trailing behind. With promised customisation options and the ability to train on specific characters or environments, Runway Gen-3 is poised to revolutionise AI-generated storytelling once it becomes publicly available shortly. OpenAI better wake up and release a competitive model like Sora soon, or they risk being left behind in this rapidly evolving field.

[@aisamsonreal]

It certainly does Dophler! Thanks for all your support on this journey

[@I-Dophler]

@aisamsonreal Thank you for your kind words and support. I sincerely appreciate your presence on this journey as we explore the incredible potential of AI technology together. Your encouragement fuels my passion for creating valuable educational resources that empower us to harness this revolutionary force for the betterment of humanity responsibly. Let us continue walking this path, learning, growing, and shaping the future collaboratively.

[@lionhearto6238]

UFO sightings about to be at an all time high

[@aisamsonreal]

I just had an alien to breakfast

[@EscapeStoriesMix]

it seems nice, but it would be nice to watch a comparison using the same prompts with different AI tools

[@aisamsonreal]

Yes, I observe a comparison between Luma in the video and I'm planning a big rundown soon

[@zeztox]

This tech is moving super fast :o

[@melissaweyrick5311]

congrats on hitting 100k 🎉🎊🎉🎊👍

[@aisamsonreal]

thanks for your support!

[@chouchfroukh]

100 k subscribers. You deserve more.

[@aisamsonreal]

I appreciate that!

[@I-Dophler]

Runway Gen 3 looks incredibly promising for generating high-quality and realistic AI videos. The ability to maintain consistency and create detailed scenes is impressive. I look forward to exploring its capabilities.

[@kareemebaid]

Why can't I see it on my account? i am still on version 2

[@I-Dophler]

@@kareemebaid OK.

[@freehaven-junprince2376]

This is pretty exciting for people wanting to make music videos. Not sure about full length story telling, but there’s a lot of potential for people with premiere and after effects skills not to mention other ai skills that can be used in conjunction with this

[@bi1iruben]

So good that at 00:01 sunlight transmits straight through a solid head. Basic ray-tracing physics still need to be improved...

[@Cory-v4w]

Ai art is the medium. I am the muse. I use pleasure and desire indirect(high-context). We are here to create the multiverse & study how to combine the four fundamental forces.

[@Nailsoupmedia]

this is huge!!!

[@omnipop4936]

Why are some of the smaller dark fish on the pathway at 5:13 just disappearing?

[@blair821]

Love it.

[@CorazonDesignsandstories]

Very interesting ❤❤❤

[@omi.3D]

That's insane what AI allows us to create! Thanks for this video 🔥

[@swarlly]

I like how we flip flop from Runway is mid to Runway is the best, and back again.

[@AJONFILM]

So sick! When is that coming out? I already paid for Runaway Gen 2...

[@stevejakab]

Hi. Where’s the beta link you referred to towards the end of the video?

[@aisamsonreal]

You can drop me an email at [e-mail]

[@Somebodythatiusedtoknow99]

How much is the accuracy of specific flora considered?

[@9thebear]

Damn, that is amazing

[@junanraihan8130]

when will it be released?

[@NoobahSteve]

we are about to get the sickest high production looking Anime

[@rehdogg]

After watching this, I might have to subscribe again to runway

[@aisamsonreal]

ha! I have some many subscriptions

[@coleenpowell]

Wow, so exciting😃

[@aisamsonreal]

It really is!

[@The-Spondy-School]

Sora killed itself by failing to deliver the goods that made the world drool. The world got tired of waiting and collectively told Sora's maker to shove it.

[@dullexe]

The moment it drops everyone will flock to it, they have the most money for the research which is why it's taking long, rather have it polished than the gimmick videos we have now, the prompt field for Sora is still the longest and intricate compared to any other company.

[@MarinaArrudaa]

Amazing videeeo!!! Like always :)

[@aisamsonreal]

Thank you so much 😀

[@TheAISkizo]

Waiting to see the results of image to video ..

[@thewebstylist]

As usual they showcase beat of the best, I created maybe 2 I would actually use yesterday after a few dozen attempts

[@Bhavi9888]

U used gen 3??

[@theunboxingroom]

hell yea

[@manumartinezkcxu]

great review, thanks

[@aisamsonreal]

Glad it was helpful!

[@XNY_Music]

How cool it is, that we get to play with these incredible toys, sorry tools, and push them to the limits of their capabilities.

[@tiffanyvelky]

Until they push us to the limits of ours

[@mainakbanerjee4035]

@aisamson can we use it for free and without watermark?

[@aisamsonreal]

it will be paid

[@DorinIonescu]

Ok, I bought the entry-level. Nice movies! (but only 2 seconds?)

[@paulfedor9828]

Be great if you did a video on how to load reference....to control it better

[@huzailijamal117]

Really excited for this. Been using Gen 2 and now this, things are moving really fast. I haven't watched the whole video but is there a mention on when will it be available and if there will be a price increase for unlimited generations?

[@aisamsonreal]

Hope you enjoy it! It says it will come in days

[@thomasproshowski7538]

Wonderful video. Can you please encourage Leonardo AI to offer this feature on their website. Thank you

[@mostfunnestchannel]

My ideal would be an AI generated 3d model which you then give instructions to as to camera movement and action, you can see the same model from different angles and tell it what to move and how.

[@starterplanet]

If you create a vid of a control room, inside a spaceship, , is there a way to make the camera not move? I tried entering "tripod shot" then "camera not moving" "camera still" but the camera still panned. On Gen-2 I had camera movement to a minimum but the camera shot always moved. Sometimes you need a shot where the camera is not moving in order to edit in a live subject.

[@camelCased]

In the distant (or not that distant) future, it could totally change 3D gaming too. Imagine that game creators don't have to create 3D assets any more but generate 3D or even VR scenes through ML models.

[@killerkally7080]

Yes but can the algorithm recreate the same exact image time after time so you can alter the colour of her hair or dress without a completely different scene.

[@LouisGedo]

14:04 That will be much better achieved via image or video to video as opposed to text to video

[@benortiz9624]

Superb 🎉😮

[@voodoochild420ai]

very cool vid great info

[@aisamsonreal]

Glad you enjoyed it

[@mygamecomputer1691]

This is neat, it’ll be useful when it can be used off-line locally. I’m prepared to pay a one time fee for the program but not a subscription where they can terminate it at will, nor an online service where I’m subject to guardrails that might be Unwarranted.

[@Greenthum6]

Not going to happen. It is a cloud-based business model and it would be stupid to give that away for local use. I doubt consumer hardware is even capable to handle this.

[@analoressallano5613]

Amazing Video :)

[@aisamsonreal]

Glad you enjoyed it

[@Aiversele]

everything changes so fast arghh

[@aisamsonreal]

it is a tsunami!

[@gu9838]

to be clear its NOT a sora killer. there are still some issues BUT its getting close. def a step in the right direction and im gonna get the unlimited plan soon as it goes live.

[@mokiloke]

It still not quite at the level of the curated Sora videos which have no artefacts, but wow full credit its astounding along with Luma, and the Chinese one, video generation has really exploded in capability.

[@makulu.online]

When the generators can generate 30 seconds / 60 seconds of footage, then I'll be impressed. I'm definitely not paying for 4/5 seconds.

[@aisamsonreal]

This one has 10 seconds and you can use the last frame of the video to start the next one so you can create infinite videos

[@makulu.online]

@@aisamsonreal yes, i just saw the part of the video where you mentioned 10 seconds, and good to know you can use the last frame of the previous video. i may give this a try, been looking for a proper video generator to implement into Linux.

[@VraserX]

Sora can generate 60 seconds of video. It most likely is even of higher quality now than we have seen before, because they are still working on getting better till release next year february after the elections in inauguration.

[@makulu.online]

@@VraserX by the time sora goes live to general public it will be so nerfed and woke im not sure it will be much good. i hear they are regulating the crap out of it to satisfy hollywood. all the actors have access and finding issues with everything, seen some actors speak about it. these alternative models will have to do. the only real issue i have is the shortness of the clips and also the number of duds you have to generate to actually get a good result... you pay example 30usd a month for x credits, but you waste 60% of the credits on footage that you cant use. they should rather make it free generation, but pay for what you download ( meaning watermark free footage that is downloaded ). so once you actually happy with a generation and want to use it, then you pay for that footage. until it gets a whole lot better or better payment models ill just sit and watch and wait... more and more options are becoming available and so pricing will drop as well.

[@paulfedor9828]

I bet if I combed through all the stock video catalogues the AI was trained on...i be yout could find the stock video clip of this girl that was sourced.

[@aisamsonreal]

That would be great! Send me a link if you do

[@goliathonscave9834]

Crazy! AI generated video is just over 1 year old yet we're already on Runway Gen 3!! What the hell will AI generated video look like in 3 years??!!

[@aisamsonreal]

it is mind-blowing

[@rustneversleeps85]

Why does everyone confuse the term 'cinematic' with 'videogamematic' these days? Is it due to all the rollercoaster movies?

[@WyattRocha]

we are cooked 😭😭😭

[@realmakebelieve]

Can you please do a video on how people are make rap songs into 60s R and B with A.I.

[@GusArtmanAI]

hi check there pricing damn i hate the credit system they use and they reset every month , for unlimited generation its a bit expensive

[@HalalVibes_Official]

In your opinion, which video ai platform is the best ?

[@maxfrompax]

No way 🎉

[@aisamsonreal]

party!

[@punchtalestudio]

I don’t see version 3 on runway

[@aisamsonreal]

it is being rolled out gradually

[@ramstrong1961]

The days of giving millions to people for learning a script are coming to an end...finally.

[@amillionpictures]

unfortunately, they only give you 10 seconds of gen -3 generation for standard and pro accounts :(

[@carloscarcia]

Will this finally be the point that Will Smith can eat spaghetti?

[@aisamsonreal]

One can dream!

[@duckyorwell9416]

Just give me the AI versions of my favorite anime's, that will be a very sick cultural moment, Imagine not having to deal with real life actors and have AI transform your anime into realistic people and environment, not losing the voice, facial and body feathers a 1 on 1 copy only one is animation the other is AI. Cant wait for the AI version of Naruto , Bleach , Death Note etc

[@BruceWayne15325]

Very exciting. It's still easy to identify as AI, but good progress. Some examples of flaws: The right breast of the tattooed woman is unnaturally constrained as if her bra were made of sharp planes instead of curves. The astronaut's feet don't move perfectly with the ground. The woman on the subway grows a nose ring as the video progresses, you can see it shining at the end.

[@Americkaus]

The stock-video business is under threat. The content-creation business is about to boom.

[@christianhenriday]

Low budget and high quality will be the norm

[@reifuTD]

My worry about that anime shot it it look like 3d animation flattened to mimic 2d animation, giving it a floaty feel and how normal is that for an out put.

[@woritsez]

The promise of AI as a kinder revolution is appealing, especially for those disillusioned by the old models of real-world administration-models mired in deception, hypocrisy, immorality, imbalances, injustices, genocides, and ecocide. Living in an era of revolution, it seems prudent to stay informed, as revolutions often bring both progress and peril in their wake. The perception of impending changes can naturally lead to fear, but the best antidote to fear is knowledge. Knowledge is the light that illuminates the otherwise darkened peripheries. So, how do image and video generation, large language models, and artificial neural networks actually work? What are their basic operating principles, the pseudo code, the real code languages, the fallback libraries and functions, the Python scripts, the graphic interfaces, and the intricate web of individual neurons? Having a fundamental understanding of these mechanisms is invaluable. How do they define boundaries between pixels and recognize objects? What are the rules, the loops, and what exactly is this "training" that enables these predictive capabilities? Understanding these aspects sheds light on the events shaping our world and altering our environment. The machinery, though often hidden in dimly lit recesses, possesses a beauty that might be sublime in its evolving complexity. Indeed, now is an opportune time to become more familiar with it.

[@masfebstudio]

I use AI generator to create many videos. But no one view the videos. My life is sad.

[@ngothianhquy7561]

Can you combine real video with AI VIDEO

[@injectionAI]

AI is about to replace artists, musicians, and OF models

[@pratiswar5977]

Initially, but over a period of time original art, music and OF type content will be more and more valued because of their rarity.

[@injectionAI]

@@pratiswar5977 I agree. I was mostly kidding

[@phantomberzerk9486]

It's funny the first 2 sounded profounded and last one lmao

[@phantomberzerk9486]

​@@pratiswar5977I think it would hard to identify but I do think there will be more competition cause people who can write but not sing vice versa will join same with art. So everything will narrow down to who the most creative.

[@alfonsorodriguez2739]

Not going to replace them all but force them to only depend on live performances. Sports and live performances will be more valued

[@WilliamAshleyOnline]

Man when those neural links get going for VR and sensory feedback gonna be some strange days........ hopefully they get that "automated exercise with muscle stimulation and split brain workouts so you can split your exercise to your unconscious mind while in the VR world... reloaded. Hopefully this gets done in my lifetime ... hopefully the nuclear war won't set us back to miss out on that crazyness and it is used for enhancing the simulation not locking us in prisons of false realities. The thing about the ocean its really not that bright. The ocean is much thicker and less light transfering than that. its more of a blue than a light white/yellow.

[@virtual_intel]

Krea can output 10 second AI vids, I'm getting better using it, but yeah RunwayML had no choice but to play catch up as Gen 1 and Gen 2 aren't that good compared to Sora and many other competitors.

[@swarlly]

Really I just want an AI that can clip a generated character, save it, then create multiple consistent angles of that same exact character without having to prompt it again, and then do that for multiple characters. In theory it should be a simple concept that requires complex execution. Once you clip a character AI should generate a 360 view of the character you can call the data from. Then when you recall that data it should pull an angle/movement of the character that you prompt for and slap it into a scene. For videos you'd have to generate multiple 360 animations. This would get very data intensive but the company could charge per character. Once a character is deleted it should dump that data from the server.

[@johncocktoasten9228]

Seems like way too many of us are not realizing just how dangerous AI is!

[@maythesciencebewithyou]

And openAI went totally silent with SORA. They don't even post any new examples on their instagram

[@mxpwr4003]

1990s "One day , we won't make difference between fiction and reality blah blah" Well .. I think it is today

[@gumvue.studio]

wating to make my movie.

[@zebromusicofficial]

thank you ground floor AI the time is now

[@michaelk6905]

I'm starting to get annoyed. Runway has now "announced" Gen-3. After Open AIs Sora, which no one I know has ever seen live, now the next tool that promises extremely good quality. And as always, there are great videos in incredible quality. Another "game changer" that is being celebrated. You can't use it yet. The quality of all video AIs that you can actually use is currently very poor in my opinion. The quality of the announcements is extremely good. Somehow it seems to me as if we live in a world of announcements and nobody notices that they are not usable. Am I being too critical? I would be interested to hear what you think of it. Sorry, I'm a bit annoyed right now 🙄😄