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

Seedance 2.5

seed.bytedance.com

Post reply on HN
Print view View original

261–270 of 278 posts

Re: Seedance 2.5

#261

Post by Nition »

galkk wrote: post #158
Nition wrote: post #75

Earlier quoted context omitted.

No longer being able to tell whether any image or video is real due to the extreme ease of creating realistic fakes, is itself a harm.

It will be solved. Walleted, vertically integrated Apple can do a digital authenticity signature and guarantee that the video you see was at least shot on iPhone, and I won’t be surprised if it will happen soon. This + some lidar info will be enabled and everybody except minority will say that this is a great thing to have.

Couldn't I just take a photo of an AI image with my phone? I suppose that'd where the LIDAR comes in.

Re: Seedance 2.5

#262

Post by sickcodebruh »

zuzululu wrote: post #66
sickcodebruh wrote: post #57

The sample video looks incredible and the ability to keep details consistent for so long is impressive but it _still_ looks screams of AI in every single shot. I can’t put my finger on why, something about the way the rooms are put together, the expressions of the faces, the movements… It’s very unsettling.

No need to fret, this is the worst it's going to be. Remember, Will Smith + Spaghetti videos?

Wasn’t that a different level of challenge? It was easy to call out the problems with it and recognize gaps to close. With Seedance 2.5, I’m not sure how I’d describe it other than “looking like AI video and all AI video is inherently off-putting”. People smarter than me are working on this but I’d be surprised if it ever actually became tolerable to look at it.

Re: Seedance 2.5

#263

Post by userbinator »

weitendorf wrote: post #60
Genego wrote: post #56

Whenever I see the new releases around video generation (and image) generation models, I get goosebumps, because it just feels so fun to work with them. But then I remember that I spend upwards of $10k on inference generating well over 50k images for storyboards, training models; and probably creating almost an hour of video (I assume). Yeah, I get that things can be economic if you don't use the latest models (ran s…

I've been dabbling in this space on the application layer and have spent a few hundred myself experimenting with video models. It can be really entertaining/addicting to build with them because you're essentially pulling the slot machine and having TikTok/Marvel/YouTube come out of it. I think that was the bet with Sora but the problem is mostly that the novelty wears off quick, and most people want to just consume c…

most people want to just consume content without typing in what content they want to see

I think that's always going to be true, but making it easier to produce will definitely increase the number of people who otherwise wouldn't bother.

Re: Seedance 2.5

#264

Post by userbinator »

bensyverson wrote: post #79
getstowly wrote: post #52

The quality of AI videos is blowing my mind. I can still see things that seem a bit "off" but it's hard to distinguish AI videos from real videos. Even blockbuster movies are starting to be dwarfed by what AI can create. How long until we see the first full length AI movie hit the theaters? No actors, no development team, just a guy prompting AI...

People said this kind of thing about MiniDV cameras, and then iPhones. The reality is that technology hasn't been a barrier for filmmakers for a very long time. Writing is still the hard part.

Or more precisely, coming up with something interesting that people will like.

Re: Seedance 2.5

#265

Post by HDBaseT »

damsta wrote: post #24

I don't think audio, image or video generation should exist. I haven't seen enough positive applications to justify the amount of harm these tools are being used to cause.

It seems like the race to the bottom slowed down with audio, video and image generation models.

OpenAI has completely shifted focus, Google still is working on various toolings/models but ultimately most people are unwilling to pay for these generations, whereby code, can easily be burned (and paid for) by businesses.

image gen isn't a proven market (at least not yet).

Re: Seedance 2.5

#266

Post by chrysoprace »

tgv wrote: post #241
chrysoprace wrote: post #101

Earlier quoted context omitted.

While I generally think it's balanced against a lot of negative applications (misinformation, deepfakes, etc.), there are some good applications. I think regulation just hasn't sufficiently caught up with the harm they can cause. To give a good use case (audio generation specifically), the 2002 game Morrowind has had a resurgence lately and it's a really great game, but it lacks fully voiced lines for dialogue. A com…

Cause chaos and let society fall into disarray, and upgrade an old game with some trifle, or not having either of those? Tough choice. > They don't take away jobs from artists In this particular case . In other cases, they do. I can't believe you think this feeble example justifies anything.

My comment addresses this.

Regulation around the world needs to catch up to address the harmful nature of the technology, as it is currently moving faster than the politicians can / want to move.

Re: Seedance 2.5

#267

Post by rhines »

galkk wrote: post #165
zmmmmm wrote: post #31

Earlier quoted context omitted.

The thing is, it has the same essential property that you don't control it any more. The problem with these tools is the lack of fine grained control means there is no room for you to express your individual creative input. Your total input is a few sentences of a prompt, then the AI did all the creative part. If the creative input is what you enjoyed, it is actually not that much more (or even less) here than it was…

It will change soon. Ai is enabler. Again, as with photography and especially digital photography- everybody can click button but not everybody is photographer. Ai helps to raise the floor, not the ceiling

It is interesting to consider what will make AI generated works great, if we reach a point where they're seen as such.

For painting, realism was really what made works great for a long time. Then photography came along and realistic depictions of things were no longer novel, so painting became about representing things in a more surreal way, and novel representations of perspective and emotion became great. Meanwhile with photography initially just capturing a clear image was impressive but now you need to capture something truly rare to have a great photo.

With AI I'm struggling to imagine what will be hard enough to achieve that it remains novel.

Re: Seedance 2.5

#268

Post by monk_grilla »

Genego wrote: post #56

Whenever I see the new releases around video generation (and image) generation models, I get goosebumps, because it just feels so fun to work with them. But then I remember that I spend upwards of $10k on inference generating well over 50k images for storyboards, training models; and probably creating almost an hour of video (I assume). Yeah, I get that things can be economic if you don't use the latest models (ran s…

Ask someone for feedback on your website. To you it might look ok, but to me and I think many others, it is unintelligible. I have no idea what I'm looking at and it is impossible to parse.

That is the primary issue, and secondarily it is a cookie-cutter AI slopfest that will make any technical person not submerged in Kool-Aid click away.

Please just get your AI to follow some basic UI best practices or pick up a UI library.

Re: Seedance 2.5

#269

Post by EspoirBao »

jjcm wrote: post #37

The quality is quite high, but an observation is the direction they're taking these models correlates heavily to the usage demand of China vs the West. Specifically, they're immensely focused on t2v for action / high effect shots. There's one human reference shot in the entire release page, and that one doesn't focus on dialog at all. For filmmakers I've talked to in the US, one of the biggest demands they have is v2…

[flagged]

Re: Seedance 2.5

#270

Post by bsenftner »

toasty228 wrote: post #258
bsenftner wrote: post #201

Earlier quoted context omitted.

The lack of creative understanding is astounding: all these generative media AI are fantastic for education applications. Sure, the vast majority want to use them to jerk off, but that's what humanity does anyway. Education is where these are extremely useful, and that is for the aid in conveying understanding to those that need the Jazz Hands of an image or audio or a video presentation to grasp the concept they are…

It's been here for a while, where are the numbers? Youtube is also great for education, yet 99% of the watch time is pure brainrot entertainment

The numbers reflect the vast majority jerking off, self entertaining. To find the 1% of the 1% that use anything for self improvement, one needs to look at the statistical tails. Because that market is so small, and their skills so advanced, you will not hear about them because they represent too small a market for our pop culture to try to leverage them for profits, other low hanging fruit is far too sweet, such as all the idiots that want to jerk off and will pay.
Post reply on HN