Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Veo 2: Our video generation model

deepmind.google

Post reply on HN
Print view View original

111–120 of 342 posts

Re: Veo 2: Our video generation model

#111

Post by noch »

handsaway wrote: post #44
noch wrote: post #10

Earlier quoted context omitted.

> Judging by how they've been trying to ram AI into YouTube creators workflows […] Thanks for sharing that video and post! One way to think about this stuff is to imagine that you are 14 and starting to create videos, art, music, etc in order to build a platform online. Maybe you dream of having 7 channels at the same time for your sundry hobbies and building audiences. For that 14 year old, these tools are available…

"remix culture" required skill and talent. Not everyone could be Girl Talk or make The Grey Album or Wugazi. The artists creating those projects clearly have hundreds if not thousands of hours of practice differentiating them from someone who just started pasting MP3s together in a DAW yesterday. If this is "just another tool" then my question is: does the output of someone who has used this tool for one thousand hou…

> "remix culture" required skill and talent.

We were told that what we were doing didn't require as much skill as whatever the previous generation were doing to sample music and make new tracks. In hindsight, of course you find it easy to cite the prominent successes that you know from the generation. That's arguing from survivorship bias and availability bias.

But those successes were never the point: the publishers and artists were pissed off at the tens of thousands of teenagers remixing stuff for their own enjoyment and forming small yet numerous communities and subcultures globally over the net. Many of us never became famous so you can cite our fame as proof of skill but we made money hosting parties at the local raves with beats we remixed together ad hoc and that others enjoyed.

> The artists creating those projects clearly have hundreds if not thousands of hours of practice differentiating them from someone who just started pasting MP3s together in a DAW yesterday.

But they all began as I did, by being someone who "just started pasting MP3s together" in my bedroom. Darude, Skrillex, Burial, and all the others simply kept doing it longer than those who decided they had to get an office job instead.

The teenagers today are in exactly the same position, except with vastly more powerful tools and the entire corpus of human creativity free to download, whether in the public domain or not.

I guess in response to your "required skill and talent", I'm saying that skill is something that's developed within the context of the technology a generation has available. But it is always developed, then viewed as such in hindsight.

Re: Veo 2: Our video generation model

#112

Post by terhechte »

carlosjobim wrote: post #101
terhechte wrote: post #59

Earlier quoted context omitted.

Back when computers took up a whole room, you'd also have asked: "but what exactly is this useful for? B-Roll some simple calculations that anybody can do with a piece of paper and a pen."? Think 5-10 years into the future, this is a stepping stone

They were calculating missile trajectories, everybody understood what they were useful for.

https://www.lexology.com/library/detail.aspx?g=164a442a-1b90...

Re: Veo 2: Our video generation model

#114

Post by noch »

Philpax wrote: post #67
handsaway wrote: post #44

Earlier quoted context omitted.

"remix culture" required skill and talent. Not everyone could be Girl Talk or make The Grey Album or Wugazi. The artists creating those projects clearly have hundreds if not thousands of hours of practice differentiating them from someone who just started pasting MP3s together in a DAW yesterday. If this is "just another tool" then my question is: does the output of someone who has used this tool for one thousand hou…

> does the output of someone who has used this tool for one thousand hours display a meaningful difference in quality to someone who just picked it up? Yes. A thousand hours confers you with a much greater understanding of what it's capable of, its constraints, and how to best take advantage of these. By comparison, consider photography: it is ostensibly only a few controls and a button, but getting quality results r…

> Just this last week, I went to two art exhibitions in Paris that involved generative AI as part of the artwork; here's one of the pieces

The exhibition you shared is rather beautiful. Thank you for the link!

Re: Veo 2: Our video generation model

#115

Post by zb3 »

the8thbit wrote: post #77
zb3 wrote: post #3

We should collectively ignore these announcements of unavailable models. There are models you can use today, even in the EU.

I don't see why, unless you think they're lying and they filmed their demos, or used some other preexisting model. I didn't ignore the JWST launch just because I haven't been granted to ability to use the telescope.

Back when Imagen was not public, they didn't properly validate whether you were a "trusted tester" on the backend, so I managed to generate a few images..

..and that's when I realized how much cherry picking we have in these "demos". These demos are about deceiving you into thinking the model is much better than it actually is.

This promotes not making the models available, because people then compare their extrapolation of demo images with the actual outputs. This can trick people into thinking Google is winning the game.

Re: Veo 2: Our video generation model

#116

Post by markus_zhang »

visnup wrote: post #109
jasonjmcghee wrote: post #45

I appreciate they posted the skateboarding video. Wildly unrealistic whenever he performs a trick - just morphing body parts. Some of the videos look incredibly believable though.

our only hope for verifying truth in the future is that state officials give their speeches while doing kick flips and frontside 360s.

Maybe they will do more in person talks, I guess. Back to the old times.

Re: Veo 2: Our video generation model

#118

Post by kaonwarb »

jasonjmcghee wrote: post #45

I appreciate they posted the skateboarding video. Wildly unrealistic whenever he performs a trick - just morphing body parts. Some of the videos look incredibly believable though.

This was my favorite of all of the videos. There's no uncanny valley; it's openly absurd, and I watched it 4-5 times with increasing enjoyment.

Re: Veo 2: Our video generation model

#119

Post by stabbles »

visnup wrote: post #109
jasonjmcghee wrote: post #45

I appreciate they posted the skateboarding video. Wildly unrealistic whenever he performs a trick - just morphing body parts. Some of the videos look incredibly believable though.

our only hope for verifying truth in the future is that state officials give their speeches while doing kick flips and frontside 360s.

sadly it's likely that video gen models will master this ability faster than state officials

Re: Veo 2: Our video generation model

#120

Post by mikepurvis »

stabbles wrote: post #119
visnup wrote: post #109

Earlier quoted context omitted.

our only hope for verifying truth in the future is that state officials give their speeches while doing kick flips and frontside 360s.

sadly it's likely that video gen models will master this ability faster than state officials

Remember when the iPhone came out and BlackBerry smugly advertised that their products were “tools not toys”?

I remember saying to someone at the time that I was pretty sure iPhone was going to get secure corporate email and device management faster than BlackBerry was going to get an approachable UI, decent camera, or app ecosystem.

Post reply on HN