4 ms·
>How could you make sure it couldn't? I just realized the answer to this is pretty obvious, you could have a network trained to classify child pornography and
by zardo 10y ago
>How could you make sure it couldn't?
I just realized the answer to this is pretty obvious, you could have a network trained to classify child pornography and use it to censor the output.
Getting the training set would be an issue, you would probably have to work with law enforcement to do it.
- NickNameNick 10y agoSurely you would just filter on the input?
- zardo 10y agoMaybe, it seems like there might be ways to work around an input filter though.
- WhitneyLand 10y agoImpossible. There is no accurate way to determine age by external appearance. For example a 15 year could look older than an 18 year old.
- zardo 10y agoIf it's being generated no characters actually have an age. Any that are plausibly adults would presumably be legally in the clear.
- DrStalker 10y agoThat doesn't stop law makers from trying. Like the mess over small-breasted porn being banned in Australia in 2010: > While the ACB claims that there is no blanket ban on small breasts as such, women over the age of 18 with small breasts who might look young ARE banned. http://www.inquisitr.com/59633/australian-government-censor-confirms-small-breast-ban-sort-of/ http://www.inquisitr.com/59633/australian-government-censor-...
- gcr 10y agoIncidentally, that's almost exactly how generative adversarial networks work. To generate realistic images, you make two neural networks: one of them (D) takes an image as input and decides whether it's real or whether it's the output of (G). The other (G) takes random noise as input and turns it into an image that will fool the (D) network. Make them fight until they both get strong, and then use (G) as the final model.