We would have thought that most of the population would not need their computer to tell them if an image is porn or not, but Yahoo seemed to think it was a killer app.

Yahoo’s open saucy software uses deep learning, which requires the computer to see a lot of porn and then if it does not overheat, decide if a new image is smut. The model that’s now available on GitHub under a BSD 2-Clause license comes pre-trained, so users only have to fine-tune it if they so choose. Like most porn it works better if you have a coffee first, oh sorry we misread the Yahoo press release, apparently the model works with the Caffe open source deep learning framework. The team trained the model using its now open source CaffeOnSpark system.

Yahoo thinks that the system will be used by those companies who are terrified that their employees are looking at things that they shouldn’t, of course parents who are terrified that their kids will see porn and learn about sex before their parents tell them about it at age 40.

Yahoo research engineer Jay Mahadeokar and senior director of product management Gerry Pesavento wrote in a blog post.

“To the best of our knowledge, there is no open source model or algorithm for identifying NSFW images.”

Of course it is not the perfect censorship system and it is not clear how it defines what is porn, something which has blighted the movie industry for a hundred years since the US introduced the Hayes Commission.