Live data from Hacker News

Decensoring Hentai with Deep Neural Networks

github.com

21–30 of 73 posts

Re: Decensoring Hentai with Deep Neural Networks

#21
post #18

I wonder what will happen when neural networks can generate illegal content, like child pornography.

It will vary by country. In a lot of Europe and the US the content will still be illegal. In Japan it won’t be. At some point the issue is going to come down to what you want to spend manpower and money fighting, people who actually endanger and harm children, or people who use computers to emulate it. It might not be a very clear cut thing though, because I frankly wouldn’t be shocked to find out that the overlap be…

But how would you outlaw it? A neural network is simply a set of coefficients, would certain coefficient combinations be illegal? It's an interesting question.

PS: I am obviously against child pornography, I'm just curious about the implications of being able to generate illegal content and how governments would deal with it.

Re: Decensoring Hentai with Deep Neural Networks

#23
post #9
post #5

I believe "Image Inpainting for Irregular Holes Using Partial Convolutions" has been superseded by "Free-Form Image Inpainting with Gated Convolution", especially for interactive use cases? https://arxiv.org/pdf/1806.03589.pdf http://jiahuiyu.com/deepfill2/ I think the most interesting part of this project would be generating the dataset. That is a lot of manual redrawing of vaginas/penises. Or the reverse, where you…

There's tens of thousands of manually decensored manga, you could download and compare them to original to identify censored regions and use as training I suppose.

Is it bad form to share one of these "datasets" here?

Re: Decensoring Hentai with Deep Neural Networks

#24

I predict that the porn industry will be the first industry to suffer mass unemployment due to AI. Phase I: Deep learning combined with realistic physics simulation will eventually replace human actors. At some point amateurs can record their movements, then use variational autoencoders to improve their features (bigger boobs, thinner waist, smooth skin) and replace professionals. This is 'Uber phase' for porn and dr…

What? Think about it, call center employees are going to suffer first on a large scale. NLP is getting better and Google Duplex already showed us that human voice is no problem anymore.

It's way easier to remove call centers than coming up with physics simulations that replace humans. Also I think porn is incredibly cheap to produce so everything must be automated to beat the current production cost.

Re: Decensoring Hentai with Deep Neural Networks

#25
post #18

Earlier quoted context omitted.

It will vary by country. In a lot of Europe and the US the content will still be illegal. In Japan it won’t be. At some point the issue is going to come down to what you want to spend manpower and money fighting, people who actually endanger and harm children, or people who use computers to emulate it. It might not be a very clear cut thing though, because I frankly wouldn’t be shocked to find out that the overlap be…

But how would you outlaw it? A neural network is simply a set of coefficients, would certain coefficient combinations be illegal? It's an interesting question. PS: I am obviously against child pornography, I'm just curious about the implications of being able to generate illegal content and how governments would deal with it.

How is cp illegal now? Each image is just 1s and 0s on your drive. It's really just an illegal number.

Re: Decensoring Hentai with Deep Neural Networks

#26

I predict that the porn industry will be the first industry to suffer mass unemployment due to AI. Phase I: Deep learning combined with realistic physics simulation will eventually replace human actors. At some point amateurs can record their movements, then use variational autoencoders to improve their features (bigger boobs, thinner waist, smooth skin) and replace professionals. This is 'Uber phase' for porn and dr…

I am going to call this the Great Stimulus.

Re: Decensoring Hentai with Deep Neural Networks

#28

Earlier quoted context omitted.

But how would you outlaw it? A neural network is simply a set of coefficients, would certain coefficient combinations be illegal? It's an interesting question. PS: I am obviously against child pornography, I'm just curious about the implications of being able to generate illegal content and how governments would deal with it.

How is cp illegal now? Each image is just 1s and 0s on your drive. It's really just an illegal number.

That's pushing it too far. An image might be represented as 1s and 0s but what it represents is a "real world event" which actually happened. If it was stored on magnetic tape instead of on a hard drive it would still represent the same thing. It's what the event represents (child abuse) which is illegal.

However, a neural network represents a mathematical model, not a specific "real world event". Although I suppose you might argue that since the dataset required to generate it would be illegal, it is illegal?

Re: Decensoring Hentai with Deep Neural Networks

#29
post #18

Earlier quoted context omitted.

It will vary by country. In a lot of Europe and the US the content will still be illegal. In Japan it won’t be. At some point the issue is going to come down to what you want to spend manpower and money fighting, people who actually endanger and harm children, or people who use computers to emulate it. It might not be a very clear cut thing though, because I frankly wouldn’t be shocked to find out that the overlap be…

But how would you outlaw it? A neural network is simply a set of coefficients, would certain coefficient combinations be illegal? It's an interesting question. PS: I am obviously against child pornography, I'm just curious about the implications of being able to generate illegal content and how governments would deal with it.

https://en.wikipedia.org/wiki/Illegal_number

Re: Decensoring Hentai with Deep Neural Networks

#30

Earlier quoted context omitted.

How is cp illegal now? Each image is just 1s and 0s on your drive. It's really just an illegal number.

That's pushing it too far. An image might be represented as 1s and 0s but what it represents is a "real world event" which actually happened. If it was stored on magnetic tape instead of on a hard drive it would still represent the same thing. It's what the event represents (child abuse) which is illegal. However, a neural network represents a mathematical model, not a specific "real world event". Although I suppose…

In most Western countries, the content can be illegal independent of the real-world event being legal or not, or even if there's no real-world event at all (i.e. fictional depictions). It's not illegal for a teenager to pose in the nude, but it might be considered illegal for them to take a photo of themselves and post it on the internet.

I think linking illegal depictions to illegal events would be saner, but that's not how it currently works.

Post reply on HN