I recently subscribed to some AI video-creating software, and OMG, it's amazing for making pie scenes.

You can use subjects such as people in scanned photos, yearbooks, and social media! (Celebs aren't allowed).

I created the attached pie video simply by having AI pick a made-up image of a woman, and I feed the app some details and instructions.

It's Google Whisk. It takes some experimenting, but I think I have the hang of it!

The fact that it's AI and not real women being pied defeats the entire point of wam, IMO. There's no humiliation if there's not a real person involved. This new timeline humanity seems to be on is exactly why movies have always depicted AI as malevolent and impersonal. I'd like to be moved to the timeline where humans are still the ones making the pie scenes.

Good for you. Personally AI is soulless echo chamber garbage. I enjoy the art of WAM and believe it needs the artist for that. I also prefer WAM that doesn't contribute to water and land pollution from massive AI data centers.

I believe that you believe that. But the physics, the synching, the emotional expressions, there's a lot that's not there yet. It just isn't.

And using a mainstream company's app to generate videos of women in any kind of context that The Machine could consider kink? Good luck doing that long term.

Fuck AI.

This 100%
I'm a big fan of AI from a technology perspective and have spent a lot of time training models. Image models are getting close to real but video still has a way to go. Sora2 very good and is miles ahead of VEO3 and whisk and still have some terrible physics issues. I've no doubt things will improve more quickly.



I find the idea of using real photos quite disgusting and non consensual. The technology for deep fakes is scary and it's going to get worse.

I don't see an attachment

i've never been one to make fun of someone else's fetish, but the "nonconsensual" thing is a very good point. and after creating the model, assigning a personality, adding in all the other details, in the long run, wouldn't you just be playing with yourself?? keep it "real"..

I can see the Terminator franchise coming true when AI gets fed up with all of our stupid AI requests. I know I'm in the minoriuty but I hate AI with a passion. it's lazy, it's nothing new and it can't think for itself. There is no unqiue aspects of it.

AI won't need an army of robots to conquer us. We're going to happily give it everything it needs.

The idea of there not being a real person involved doesn't bother me, but then I can be entertained by a fictional WAM story and there aren't real people involved in those either. And there are scenarios I'd generate with AI but wouldn't want to see performed by real people, if they were unpleasant or even dangerous for those involved.

That said, a lot of the AI-generated videos I see on YouTube at the moment are situations where the same thing already exists, done better, with real people, and some of them definitely feel like cynical low-effort attempts to grab attention from people who've come looking for the real thing.

Oh joy...I'm going to get yelled at again.

Ok,, so 3 things....

First, I agree nothing is better than non A.I. Wam.

Ok, glad we got that out of the way.

Second, (and I have asked this before)

What happens if I take a photo of a WAM session from say 20 years ago of which there is no video (but it actually occurred, hence the picture) and then toss it into an AI generator?

You have "real" WAM because it did actually happen with real people but it's still created synthetic technically.

Third, what if I take a real live gorgeous model and create a video where she gets pied by.............an A.I. model?

Then, what do we have?

Robby

I expect your papers on my desk in the morning.

I use Topaz to remaster my old videos. It's an AI upscaler and I think it does a good job.

Outside of that I'm not a fan of AI generated splosh videos. Some of them are interesting but not erotic.

don't fucking do this you creep

Not there yet but one day it will be awesome what can be created.

That's whole wider question than just WAM. I go back to my poit about Sora2. It is by far the best AI model for both WAM and nonWAM videos. The first couple of days it was released, the guardrails and filters were very low. It was very easy if you were inclined to create videos involving celebrities. OAI clearly saw the potential dangers on launch and prevented you from uploading images of real people to involve but it didn't prevent you prompting for celebrities.
Unsurprisingly, filters have increased significantly around celebrities and other IP. Specific dead celebrities such as MLK and Robin williams are been blocked due to the amount of abusive meme videos.

I guess my point is the technology has moved significantly just in the last month and whilst this is really only available in good quality online, it's only a matter of a few years before local models reach that level. I class WAN2.2 or 2.5 as good enough yet.
From a moral perspective, as long as consent is given, I don't care whether a video or image is AI. I'm a big supporter of AI and whilst it's never going to be the same as real life, I can see it having it's place.
Whether someone likes or dislikes AI isn't my concern and I won't waste my time trying to convince them to change their mind.

I look forward to getting vaporized by my modem after going over my government allotted monthly bandwidth limit

Think of AI as your yard. You don't need grass or plants, but you will spend a great deal of time and money maintaining it. It doesn't give you anything in return, but you do your best to keep it healthy. Thats AI. You don't get anything of value, but you'll continue to feed it.There is nothing AI offers that you can't do yourself.

Having initially been enthusiastic about AI in WAM, I am starting to become, well, less enthusiastic. This is both on the grounds of the limitations of the technology and on the ethical issues. The potential for abuse in creating fake footage of real people is concerning - not just a lack of respect for deceased celebrities, but people have been wrongly convicted on misidentified recordings, and now we make public the ability to create videos with convincing likenesses of real people and animate them. Even with sophisticated filters on well-known celebrities and public figures, any 'image to video' feature will have no real mechanism for detecting if the source image is being used with consent.

The issue that follows here is that humans in AI are now very lifelike, well beyond the 'six fingers' era, as is much of the animation, but at the same time the actual outputs rely on training data it can imitate and as a result we are currently in a state where what can be made is relatively limited and then only if the LLM can interpret the prompt as the user intended it. In the niche sphere of WAM, all the successful gungings start to look the same after a while. There's a particular look to AI videos in general once you see enough of them in which the same basic scene just repeats, and not just because users are unimaginatively imitating each others' prompts and ideas.

The real kicker is that currently the entire industry is being propped up by investor money because it in turn is the only thing generating growth and keeping the US out of recession. OpenAI, according to several sources, loses around $5 on every Sora video and everyone needs to keep the cycle of hype going. But they can't simply charge users the actual cost since the tech industry model is to saturate the user base and create a monopoly. Plenty of people warning that it's a bubble akin to dot.com of the early 2000s.

These have two different answers in my book. Each considers the use of A.I. differently.
If you are tossing the picture, video etc into an A.I. generator to clean up the picture or the footage that is a more acceptable use for A.I. You are not relying on A.I. because you can't do it yourself. You are using A.I. to help with a restoration of something that was real and did happen.

Taking a real model and tossing her into a green screen or similar situation to try and record real reactions to something that is not happening is fake as hell and will look it. Just like all those bad TUBI movies that rely on A.I. for most of it. What is the point of making the video if the model is not getting pied? As long as your are not trying to pass it off as real that is what matters. The deception factor is high with a situation like this. To me, using A.I. and trying to pass it off as real is no different than Chantelle's or Jessie's boyfriends using their account to post as them.

It's hard to actually do the 'scenes' you like to do, what you're 'into', in real life..... Think of the mess, the 'clean-up' afterwards..... Happy to have had some really good 'real-life' experiences.....


And that's where 'A.I.' comes into it..... It's a 'fantazie', you create what you like......

If you don't like 'A.I.' pixs or videos, then just don't watch 'em......

I was dead serious, I hate ai and I think it should be obliterated. I genuinely think my comment will be a real thing in the future. Perhaps it's a bit doomer-ish, but I have a sneaking suspicion that tech oligarchs don't have our best interest in mind. That being said I'd rather get vaporized than be forced into the Amazon Essentials brand lithium mines.

Annnnnnnnyyywayyyyyyy.

Thank you to you, though, John Q. Pie-Enjoyer for reading my ramblings, my pie in the face kink porn videos are available for purchase below.

Also consider this. All these fucking data centers consume an incredible amount on electricity, so much so that's its increasing OUR power bills. My electricity bill has increased 100% over the last year and my consumption hasn't increased at all.

Have fun making us all go broke over your fucking boner, dummy.

to be fair, electricity just isn't something that should be a problem in any developed country. it's entirely a problem of our own making from decades of opposition to nuclear power and renewables. we could easily have energy abundance if people stopped opposing technology they don't understand just because it sounds scary.

Yeah, don't use AI... the more people post AI junk on here, the less people will post real content!

Yep! And research is suggesting it will take roughly $2T to build out the infrastructure in the US to support demand, with the major AI companies expected to have an $800b shortfall just from dealing with that expense, let alone turning a profit, covering other expenses, training models, etc.

There is so much AI stuff out there now but it just looks AI and defeats the object of WAM. It will slowly kill off people producing their own AI with REAL people which for many is the point of wam. It's knowing that it's a real person being gunged or slimed. AI will become so oversaturated that whatever you want will be there but as said, the one thing missing is the reality of it being a real person.

scooterpie17 your post might be more well received and discussed on the ai wam group:
https://umd.net/groups/group/ai-wam

You mentioned an attached video, but umd does not allow videos to be attached to forum posts. You will need to convert it to a gif, or upload it as a video under your profile. There is a rule in the ToS that ai videos must be at least 5 minutes. If this post was to share ai content, then you should have checked the synthetic checkbox.

Regarding your comment about the subject - you can create this content privately, but do not share it on umd unless you get permission from the person, and the person you are creating ai content of consents to you using their likeliness. There is a rule in the ToS about this:
https://umd.net/termsofservice/#deep_fakes

I'm not sure that's necessarily true. I think some people will always want "real people" wam, because of that realness - knowing it actially happened. With AI you'll be able to create any scenario you like, but as others have said, it's effectively a hyper-realistic animation, a cartoon, albeit a very realistic one.

For some, the realistic visuals will be enough. But for others, knowing it happened for real will be paramount.

As I see it, the WAM marketplace will split into two sections:
Video producers selling scenes to people who want to know it really happened.
AI fetish providers selling the ability to create content to people who just want the visuals.

The thing I don't see selling is AI-generated WAM content. Why settle for someone else's fantasy if you can generate your own perfect one? There may be a market in tutorials teaching people how to use AI tools to create their perfect fantasy but those will be monetised YouTube content rather than sold by WAM producers.

Personally I plan to go on shooting real-people WAM scenes for as long as people are buying them and I have willing models. Through business connections I have access to some powerful AI tools, and it's great fun to play with them, but there is something about knowing it happened for real that does make a difference.

TBH the best results I've had with AI aren't photorealistic, but rather creating what look like 19th century oil paintings of WAM scenes - things that never existed but it'd be fun if they did. We might eventually use some of those in the "Explore Saturation Hall" areas of our site, for people who enjoy the backstory and historical characters - but the actual for-sale content will always be 100% real people in real scenes.

Here are a few examples of the painting style images - fun, but in no way pretending to be real.

I agree and disagree to this. I can't see the demand for producer content dropping in any shape or form in the foreseeable future. As you say, the fact is, there is no substitute for real wam for a variety of reasons.

AI technology is advancing fast but I still don't see it being a viable alternative any time soon. Messy is just to hard to do right, the difficulty way surpasses clean images and videos.

Commercial models just aren't trained on big enough wam datasets whether that is video or images. you may get some some reasonable success but it can't generate what it hasn't been trained on.

The guardrails on commercial models is currently set up to restrict adult, NSFW and fetish material. SORA and GPT is well aware of these situations and will block them. I can't see the future but even with relaxed adult restrictions coming on images I can't see that pushing out to video.

Open models may have more long term future but they're a generation or 2 behind Sora2 and Veo3. If and when local hardware and models can compete with sora2, who is going to train them? Out of the box, open models are based on much smaller datasets. I can count on one hand the people who have trained a WAM model and even few with any success. It's never going to be the case of uploading 5 messy images and out pops a universal model.

End of the day, I don't think producers have anything to worry about in the near/medium term and after that there is still no substitute for real messy material.