It seems more promising because people have access to it now in and they are generating videos as of now and we’ve seen that it’s actually better. We had a bad impression because Sora . This time around it seems like they might actually release it within the next couple weeks I hope
People wonder why it's so good compare to Sora, here's a hint: it's Google, it has access to hundreds of petabytes and even more, of video footage in the world. The AI knows what is possible within that library of videos, and what is not. Hence it's trained very well to produce Nice quality image and respect the physics of our world.
Also, Google has been in AI for a long time. People act like OpenAI invented it while many of the world-changing papers were released by Google scientists.
@@IvanVazquezS AI is soulless, abnormal, not fun and extremely lazy. I bet people who use AI to do things instead of having fun with making stuff BY THEMSELVES are 698LBS who haven’t stood up for the past 3 years 💀
This is the first AI video i've seen that isn't just trained on a bunch of descripted videos I think this looks good for 2 reasons 1. It's google obviously, petabytes of training, But I think this is lacking cuz for sure, a bunch of it is Junk/recycled 2. I think they used a lot of specialized/expert AI, ie, physics informed AI, Clothing AI, Skin AI, foreground/background AI Ensuring that the background doesn't get mixed/becomes the foreground in just 5 seconds of blur
Hey can you tell us what you did to get early access? How did that happen? Edit: Also how long did these generations take on average? I expect it will increase/worsen once more users hope on, so i'm using it as a benchmark.
Yep, this might be probably it, guys. Once this is out, this is like the future for AI filmmaking, for films & cinema. And we can use this to make our own films
It's something incredible.. How can computer graphics programs still be important? There is no point in learning complex programs when something like this AI. exists....which is just beginning. Just for hobby maybe..
They just trained their model on billions of youtube videos and people are happy by watching this, but it will be a critical problem in future, who owns the generated content, all of those people are stupid who are making real content 😢
Damn, imagine the future of Netflix and RUclips is probably gonna be that you just tell an AI what type of content you wanna see and it will generate it uniquely for you on the fly!
Yeah but then with Neuralink or similar they’ll just understand what you want on the go so you don’t even have to type or say anything, it will just feel your vibe and what you wanna see next to be impressed and you’ll see it
ha ha ha Sora also shown lots of promises in Demo but when we finally put our hand it was both late and OK like any other AI tool. Lets see what Veo 2 do when it come sin our hand . Till then its all hype .
impressive, cant wait to get my hands on this video ai
Hope its this good once its released in 6 months
@@BrolinBrolinBrolin I wonder if it will be even better!
No one trusts these companies anymore😂😂
It seems more promising because people have access to it now in and they are generating videos as of now and we’ve seen that it’s actually better. We had a bad impression because Sora . This time around it seems like they might actually release it within the next couple weeks I hope
@@tjo6252 They gave several creators access to test it and it seems very good from their reviews comparing it to other models
If it releases in 6 months
People wonder why it's so good compare to Sora, here's a hint: it's Google, it has access to hundreds of petabytes and even more, of video footage in the world. The AI knows what is possible within that library of videos, and what is not. Hence it's trained very well to produce Nice quality image and respect the physics of our world.
Google's access to RUclips just changes everything for them.
Also, Google has been in AI for a long time. People act like OpenAI invented it while many of the world-changing papers were released by Google scientists.
They have advantage owning RUclips but sora will catch that up
it's starting to get really good I must say
Probably the strongest tool so far.
Insanely good!
I'm so impressed with this tool!
Lies
Impressive, very nice... let's see Paul Allen's AI demo
Can't wait for the ai generation to be implemented in games to make them look real
-💀🤓🤡💩
@Anon00074can you elaborate?
@@IvanVazquezS AI is soulless, abnormal, not fun and extremely lazy. I bet people who use AI to do things instead of having fun with making stuff BY THEMSELVES are 698LBS who haven’t stood up for the past 3 years 💀
My god..... I don't know if it's AI or not anymore... This is dangerous... But amazing
No it’s not amazing. AI is soulless, boring and not fun to be entertained of, in life you must embrace traditionalism and that is it
@Anon00074 You likely already see AI-generated content everyday without noticing.
@@sirklatt nope that maybe you 💀
@Anon00074 this vid is AI generated you muppet
@@sirklatt And you should be aware of this you insane tard 💀
crazy ! can you test "roller coaster" pov ? every other video generator screw that completely
Working on a video for Go-Pro style footage that should have a roller coaster one :)
And the class Will Smith eating pasta test!
@ check my X account for the spaghetti video:
x.com/jerrod_lew/status/1868809004400754871?s=46
@@jerrod_lew Incredible. Did you not prompt Will Smith? Or you did and it gave that output? I'm don't use twitter, but saw the vid.
Можешь пожалуйста снимать больше примеров и показывать? очень интересно на что она способна
Just amazing
-🤖
have you tried giving it a dialogue. can it produce a lip synced version?
Smarter decisions start here-with AI 💥
This is the first AI video i've seen that isn't just trained on a bunch of descripted videos
I think this looks good for 2 reasons
1. It's google obviously, petabytes of training, But I think this is lacking cuz for sure, a bunch of it is Junk/recycled
2. I think they used a lot of specialized/expert AI, ie, physics informed AI, Clothing AI, Skin AI, foreground/background AI
Ensuring that the background doesn't get mixed/becomes the foreground in just 5 seconds of blur
Hey can you tell us what you did to get early access? How did that happen?
Edit: Also how long did these generations take on average? I expect it will increase/worsen once more users hope on, so i'm using it as a benchmark.
@@jmg9509 I was on the waiting list and got in earlier this week.
Each generation takes 2-3 minutes and you get 4 options.
@@jerrod_lew Very, very nice. I don't think any other video generator offers 4 options like that. So far as I know, only image generators do that.
@@jerrod_lew i am waiting on the veo 1 and 2 waitlist for over a year now :(, maybe its because i rickrolled google when i put my portifolio?
Yep, this might be probably it, guys. Once this is out, this is like the future for AI filmmaking, for films & cinema.
And we can use this to make our own films
Creativity for everyone!
Nah, veo 3 probably
Ww1
Like it's going to be readily available to everyone...
I am wondering how they'll tackle continuity of say designs or characters. Hoping I see this stuff come to fruition.
It looks like all the actors in Hollywood going to be jobless 😂😂😂
Still a tiny bit uncanny, but definitively getting "there".
Needs consistent characters/environments, scene extend, story board, and character voice generator with lip sync.
Absolutely! This would be incredible.
This will be a fantastic tool inside hands of dictators...
??
It's something incredible..
How can computer graphics programs still be important? There is no point in learning complex programs when something like this AI. exists....which is just beginning. Just for hobby maybe..
I think the ones who have the technical knowledge will always have the advantage on how to use these tools.
Best.❤️❤️
다른부분은 큰 이슈가 안됨. Camera control 기능이 핵심임.
They just trained their model on billions of youtube videos and people are happy by watching this, but it will be a critical problem in future, who owns the generated content, all of those people are stupid who are making real content 😢
This is a huge challenge, I don't really have an answer yet :(
Walking and running are still problematic for even this system.
i have the same issue.
Damn, imagine the future of Netflix and RUclips is probably gonna be that you just tell an AI what type of content you wanna see and it will generate it uniquely for you on the fly!
And P-hub😂
Yeah but then with Neuralink or similar they’ll just understand what you want on the go so you don’t even have to type or say anything, it will just feel your vibe and what you wanna see next to be impressed and you’ll see it
@@tschedd Even scarier 😭
OOOOMMMMGGGGGGGGG
LESSGOOOOOOOOOOOOOO
-🤖
Soon the episode from Black Mirror "Joan is awful" will be real. Holy moly. This is getting scarier guys 😱😢
I need to see this episode!
Google aint messing about!!!.
ha ha ha Sora also shown lots of promises in Demo but when we finally put our hand it was both late and OK like any other AI tool. Lets see what Veo 2 do when it come sin our hand . Till then its all hype .
What do you mean? From my understanding, there are already people testing it, like the person who posted this video
Bye bye Sora right ?
Sora bros coping
I can do this with minimax also..
So this was GTA6
the only one that looked fake was the dude running on the track he was moving too slowly
Yeah I agree, though just loved how good his running motion looked.
This is kinda scary
Yep!
Soon video games will be generated and not rendered.
This actually excites me!
Really idont blieve it
This isn’t good
you isnt good
Billions of hours of youtube videos and metadata was used for it's training.
It really shows as well!
So it understands countryballs and jokes?