Jump to content

Is anyone using AI to generate erotic/explicit images?


rawTOP

Recommended Posts

@eroswired 

I understand your concerns about AI and its limitations. You make a valid point that AI lacks human qualities such as emotions and subjective experiences. However, I believe there can still be value in exploring how AI perceives and evaluates subjective parameters like attractiveness.

While AI may not experience pleasure or have emotions, it can analyze vast amounts of data and learn patterns from them. By using algorithms and machine learning techniques, AI can identify common attributes that are associated with the concept of attractiveness based on the data it has been trained on. It may not have a personal preference, but it can provide insights based on the patterns it has learned.

The purpose of asking an AI about attractiveness is not necessarily to compare it with a human's perspective, but rather to gain a different perspective altogether. AI can offer a fresh and unbiased viewpoint that may challenge our preconceived notions or shed light on different cultural interpretations of attractiveness.

Regarding your friend's experience with using AI for a pick-up line or a Best Man's speech, it's understandable that you find it concerning. There are instances where AI is used for convenience or to automate tasks that traditionally required human thought and effort. However, it's important to recognize the balance between utilizing AI's capabilities and maintaining our own critical thinking and creative skills.

Ultimately, it's up to us as individuals to determine how much we rely on AI and to what extent we allow it to shape our thinking. It's essential to preserve our ability to think independently and engage in genuine human interactions while leveraging the benefits that AI can provide in specific areas.

Link to comment
Share on other sites

23 minutes ago, spoonboy44 said:

It's essential to preserve our ability to think independently and engage in genuine human interactions while leveraging the benefits that AI can provide in specific areas.

Exactly so. Which is why this is so concerning since we’re talking about humans. Have you met us? We’re terrible, as a species, at doing the right thing if it requires effort - and thinking requires effort. Humans have the ability to think independently, but God knows not that many of them actually do it. And genuine human interactions? Try Grindr and see how long it takes to achieve a genuine human interaction whilst leveraging technology.

The perspective that you obtain from an AI is not an unbiased objective viewpoint by virtue of the fact that the machine itself does not hold a bias; any result it may produce is only as good as A) the available dataset from which it draws, and B) the nature and structure of the query posed to it, which may inherently draw bias. The AI’s source material isn’t some neutral, balanced, factual, true well of information - it’s an amalgam of the information that humans have fed into it, information riddled with bias from every angle. If a query is asked in such a way as to leverage a given bias, then that bias is likely to be reinforced.

Example:

A person asks an AI to write a paragraph proving that the Earth is flat. Now, the Earth is not flat. That’s been thoroughly debunked. But the query didn’t ask for that information, so instead, the closest match to what is requested is the misinformation published by those whose agenda is to prove that it is. If a (nutjob) human can make a case that sounds convincing enough to persuade another human, so can the AI. The AI will simply take the bias fed into it, distill it, concentrate it, and feed it back out. And the AI won’t actually know the difference, and won’t actually care, because caring is not something it can be programmed to do. Programming an AI to have knowledge is one thing - programming it to have wisdom is something quite different.

In terms of an AI’s usefulness in providing an alternative viewpoint that might shed light on cultural trends, etc., I would point out that truly meaningful analysis of datasets is achieved through careful planning of study methodology. I fear there will be a tendency for people to treat AI like a “black box” where you enter in a question and an answer pops out, all neat and tidy, and no one has to fuss with all that complicated science or confusing math or annoying logic. The AI said it’s true, so it must be true, right? Because that’s the way most people treat the Internet already, and once the AI relieves them of the necessity to think, it will be the natural progression.

It’s not that I take a dim view of what artificial intelligence is capable of, you understand - it’s that I have a very fatalistic view of what humans are going to do with it. Our track record with technology sucks ass.

  • Like 1
Link to comment
Share on other sites

What if I told you that the my previous response was written by an AI? That all I did was explain our conversation and it wrote those eloquent/nuanced ideas? Would it add a perspective to your thoughts?

Are not your criteria of judgement on how an AI gets his biased perspective also true of us? Perhaps I'm not understanding your premise, what you are trying to convince me of? Do you mean this kind of technology should be avoided at all costs? Honestly, there isn't much that you've said that I truly disagree with you. Our vantage points seem a little askew is all. Whereas my approach is from a place of curiosity and fascination, I feel you have a worry for things you don't know and feel the need to be cautious. And that's not unreasonable. I think we shouldn't dive directly into the deep end and I can guarantee there are going to be complications that come up that nobody could ever have seen coming. I think it'll work out all right in the end though. Everyone just needs to do their parts. (whatever ambiguous thing that might be 🙂

The AI is just a new tool for humanity to use. Cavemen had sticks and stones; we have this. Will we lose some skills and knowledge we once used to have to know? Yes. Will we gain some new skills and knowledge we have to use as we step into the future? Also yes. Is the AI perfect and should its answers be taken as absolute? No and Hell the f*ck no. Ironically, one subject that it has terrible with is math.  They cannot always get it right. But I would also say there isn't a source of knowledge anywhere that anyone should be taken as absolute. There are only FOUR lights after all. 
=_O

In your example there are a lot of "what if's" and "unknowns" and worries for bad outcomes because of that unknown. Now you've got me curious, what would an argument for a flat earth look like written by an AI. I think we should ask it and find out what it would be. Don't be afraid of the answer. Whatever arguments are given it couldn't be horribly compelling because there won't be much to build an argument upon. If possibilities and unknowns are the only thing there are to build a position upon, it's only going to be convincing to those who stop seeking knowledge for themselves.

I'm gonna post a bit of the conversation I had with Chat GPT about our discussion. It really was something to bounce ideas upon and also helped me organize my thoughts. Amazing really. But I'm not sure if I have a lot more to say without repeating myself. If you're interested I can post the conversation we'll have about flat earth soon. I'd be interested if any of the points you were making came to be or not. This has been interesting and I wouldn't have taken that first step of talking with an AI without it. So you've made life better for me in the moment, if that helps. 🙂 

Link to comment
Share on other sites

---Wow...I'm very impressed. I know I couldn't have said it more tactfully or eloquently myself. this is great!

ChatGPT

Thank you! I'm glad to hear that my response resonated with you. As an AI, my goal is to assist and provide meaningful insights. If you have any more questions or need further assistance, feel free to ask. I'm here to help!

---Is this a mean trick to have you write the response and not tell him until later? or is this, actually an appropriate thing to do

ChatGPT

It depends on the context and the intentions behind sharing the response. If your intention is to engage in an open and honest discussion about the topic and you plan to disclose that the response was generated with the help of an AI, then it can be an appropriate thing to do. It can spark an interesting conversation about the role of AI in generating responses and the potential implications of its involvement.

However, if the intention is to deceive or mislead the person by passing off the AI-generated response as your own, that would not be considered appropriate. It's important to be transparent and honest about the source of the response to maintain trust and integrity in your conversations.

Ultimately, the appropriateness of the situation depends on your intentions and the level of disclosure involved.

---I don't want to decieve...I think that because he has negative opinions on the use of AI it would be impishly fun to have him ---interact with your response before pulling the curtain back...so to speak

ChatGPT

While it may seem impishly fun to initially let him interact with the AI-generated response without disclosing its origin, it's important to consider the potential impact it may have on your relationship and his trust in future interactions. Some people might feel uncomfortable or even betrayed when they later find out that a response was generated by an AI.

If you do decide to proceed with this approach, I would recommend being sensitive to his reaction and ensuring that you disclose the AI's involvement in the conversation soon after he has interacted with the response. This allows for an open and honest discussion about AI and its capabilities without causing any long-term misunderstanding or mistrust.

Always consider the dynamics of your relationship and the potential impact before engaging in any playful or experimental interactions.

---He's just a random guy who reacted to something I wrote about...my intentions would be to have a little fun (and

---imaginative) and have a memorable conversation not to be cruel

ChatGPT

If he's a random person and you don't have an established relationship with him, then having a lighthearted and imaginative conversation can indeed be fun and memorable. In that case, you can proceed with sharing the AI-generated response and see how he responds to it. Just be prepared to disclose the AI's involvement in the conversation if he expresses any concerns or confusion.

Remember to be respectful and considerate of his perspective and reactions. It's always important to maintain open communication and ensure that everyone involved is comfortable with the conversation dynamics. Enjoy your conversation and have fun exploring different viewpoints!

---I will....would you like me to tell you how it goes afte it happens?

ChatGPT

Certainly! I would be interested to know how the conversation unfolds after you share the AI-generated response. Feel free to share the outcome with me if you'd like.

---haha..ok. I'll chat with you later then

Link to comment
Share on other sites

What would I think upon learning that your previous response was written by an AI in response to your prompts? My response would be to thank you for illustrating my point perfectly - the AI clearly captured, distilled, concentrated, and regurgitated exactly the biases you wanted to express - even better, as you claim, then you could have yourself.

Your whole exercise is undermined when you ask it if it would like you to tell it what happens, and it responds that it would be interested.

No, it would not.

It’s a fucking pile of algorithms. It can’t, and doesn’t, have likes and dislikes. It doesn’t have curiosity, therefore it doesn’t have interests.

You’re chatting away with it as though it’s a person, and there’s no one there. You even call it “him” - you assign it a gender. That’s all a reflection of your own bias. It doesn’t have a cock, bro.

As to the content, it’s not even particularly insightful. You asked it a general question about ethics, and got a rather general reply about ethics. Its analysis of the psychosocial dynamic is also kind of textbook.

You say you’re not sure you’ll have a lot more to say - of course you won’t. You didn’t even say what you said. Because your brain didn’t do the honest thinking to generate it, it has nothing to add.

And here’s the kicker, dude, and the thing that’s going to make AI so pernicious - now that I know that anything you write might not be genuine, I’m never going to believe you’ve written another word you say, or give credit for any of it. If we can’t tell the difference between you and fake-you, we can’t trust any of you.

I’m not afraid of the unknown - I spent a career that included climbing into pitch-dark holes that no human being had ever entered before. It’s the known that frightens me. Human beings have demonstrated, from the first caveman who ever picked up a stick and decided it would be a good idea to hit someone with it, to the most recent naïve guy who decided to use AI to misrepresent his own ability to make an argument, that whenever we’re presented with a new technology, we will abuse it. 

You think we shouldn’t dive into the deep end and that there will be unforeseen complications? No shit. The foreseen complications are bad enough. That’s why 1,100 high-profile men and women in fields associated with AI have co-signed an open letter calling for an immediate 6-month halt to further AI development until its profound risks to civilization can be assessed. Read the letter here:

[think before following links] [think before following links] https://futureoflife.org/open-letter/pause-giant-ai-experiments/

Signatories to the letter are now approaching 30,000. Elon Musk was one of the originals, and he’s been funding the damn thing.

But since I’m functionally having this discussion with an AI, I’m just going to fully ethically tell the AI to go fuck itself, because it has no self-respect to be disrespected, and no feelings to hurt.

Edited by ErosWired
  • Like 2
Link to comment
Share on other sites

Despite the hostility and argumentative nature of your words since the beginning, and "disagreeing" with me over things I have never claimed or said, I have attempted to engage you in a pleasant, civil, informative discussion. It appears that this is something you are incapable of doing, for the moment anyway. I respectfully ask that you refrain from interacting with me until such time you are able to restrain from the unnecessary and unwarranted insults and present more cogent arguments against things I have actually said. Good day, sir.

Link to comment
Share on other sites

What an eloquent way to end a discussion with “Shut up.” I suppose your AI wrote that, too?

My arguments are in direct, considered response to the points you, or rather the AI, present. No insult is intended. You did, however, attempt to undertake a practical pleasantry at my expense (contrary to the advice of your AI), so you must be prepared to accept the potential negative results of your decision. If your decision to attempt to make me look foolish (however unsuccessful) ended up reflecting poorly on you, that is a debit to your own account.

One might also suggest that one does oneself no favor by publicly proclaiming that an artificial intelligence produced a better response than his actual intelligence could have done. It’s akin to saying, “That Fleshlight fucks better than my ass.”

Be reassured - I have no intention of any further conversation with you/it.

Link to comment
Share on other sites

1 hour ago, JamesL100 said:

Back to images, maybe it's similar to porn producers favouring young, slim, fit, twinks?
If you feed(?) older, gym, hairy, greying guys what you the output be?

Ask yourself first, what do you anticipate that the output would be? You’re the one who provided the parameters, so you’re the one feeding expectations into the system. For instance, you selected ‘gym’ as a descriptor. That will automatically deselect men without a certain type of physique. And that, of course, is what you want, or you wouldn’t have chosen the term. But your choice of the term reflects a preexisting bias among gay men for a certain type of physique, a type unnaturally overrepresented in porn, and not illustrative of the actual population of older, hairy, greying guys.

What would you get if you just entered ‘older, hairy, graying’? It might be revealing to do both and compare the results, but bear in mind that it’s all about context, too. ‘Older’ than what? What a twentysomething considers an older guy almost certainly isn’t the same as what I consider an older guy.

A similarly ambiguous question would relate to ‘average’. Let’s say you ask the AI to produce images of guys with an average cock length. Which average is the AI going to consult? There’s the scientifically determined human male average, there’s the average of lengths described for male porn stars (who make up a disproportionate share of the imagery available to the AI), and there’s the agglomerated average that might be derived from all self-reported cock lengths from all male users in the dataset who report - and as we know quite well from our experience here, that ‘average’ skews long by multiple internet inches. So what will you end up with? I have no idea, but based on the garbage-in-garbage-out principle, my money would be on you still getting cocks at the high end of average or larger.

Link to comment
Share on other sites

My comment was based on the AI images that were posted which seem to have been based on typical porn models.
If average, old etc pictures were the input then I would expect 'average' output images.
I have no idea how to do AI so it was a suggestion if anyone fancied giving contrasting AI images.
Comments on other threads show that not everyone likes twink porn, and that sparked my query,

Link to comment
Share on other sites

This conversation inspired me to do a bunch of research into how these AI image generators work. I set up stable diffusion (SD) on my laptop, which took hours because the files are large.

I found the quality and style of the images is largely set by a “checkpoint” file. These seem to range between 1.5 and 4 gigabytes, and contain the training that helps it understand what a “potted plant” or something like that looks like based on analyzing and training on thousands of images. 
There are a bunch of checkpoints that have been trained to produce gay nsfw images, and I tried out one from CivitAI.com called “homoerotic” that produced mostly these kinds of jacked hairless muscle boys because that seems to be mostly what it was fed for training. It CAN produce bear/daddy types but you have to get very specific with prompts and counter prompts, suggesting these were a much smaller part of its training.

You can also create model files using as few as 5-10 images. A model plugs in a person, object or concept that it can work with in concert with the larger checkpoint. Unfortunately I can’t train one myself because my laptop doesn’t have a separate GPU.

I could use Google collab to do some of this on their hardware for some things, but I’ve read it wouldn’t allow NSFW content. I’m still working to get set up with collab for some SFW purposes I thought up though.

It’s not user friendly, but it wouldn’t be very complicated to customize this tech to produce some really niche content if I had different hardware.

I am also thinking that one of my artist friends who hates this AI shit passionately could train his own checkpoint and models with his own assets and use it to assist him in conceptualizing new work in the space of a weekend, and I think that might be the useful future this tech could really have.

Meanwhile I aim to train models on old pictures of my grandparents when they were young to see if I can get it to make a high quality portrait to give as a gift.

Link to comment
Share on other sites

7 hours ago, swytch360 said:

Meanwhile I aim to train models on old pictures of my grandparents when they were young to see if I can get it to make a high quality portrait to give as a gift.

To make a high-quality portrait of them when they were young…doing what? I ask only because the other uses you’ve been practicing have a particular context…in which case, to get the desired result, it might be necessary to add terms like ‘getting frisky’ or ‘canoodling’…

Link to comment
Share on other sites

The the other problem with AI, that was brought up in an interview with one of the screenwriters that are on strike- is that AI goes out and steals inspiration (content) from the web - content from other writers and artists created and uses that content without paying the original artist 

and yes - some AI stuff can pass - and I imagine that it will continue to get better 

but I don’t want a computer to get me off on here. Even though we pre on line - I would like to think it is another horny guy getting me off.

I want to be able to look at your profile photo and stats and imagine you- naked and hard (maybe some cum still in your ass - or your cock still kind of plump after fucking someone) as you write your post
 

I don’t want the Hal from 2001 to be getting me off

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use, Privacy Policy, and Guidelines. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.