Stay on target A West Virginian teenager taught artificial intelligence to generate nude portraits. And this is why we can’t have nice things.By feeding thousands of nude portraits into a generative adversarial network (GAN), Robbie Barrat developed some of the most disturbing art I’ve ever seen (and I’ve been to the MoMA).GANS are a class of artificial intelligence algorithms used in unsupervised machine learning; it pits two computer systems—a generator and a discriminator—against each other to achieve a certain goal.Invented in 2014 by Google Brain Team member Ian Goodfellow, the dueling-neural-network approach was employed last year by NVIDIA in an attempt to produce human profiles using famous faces.The project, unsettling as it was, yielded more success than Barrat.“The resulting portraits are not realistic,” the 18-year-old admitted. “The machine failed to learn all of the proper attributes found in nude portraits and instead has fallen into a local minima where it generates surreal blobs of flesh.“Is this how machines see people?” Barrat wondered.As mushy jumbles of beige against darkly distorted walls, with spilled-paint-like splotches for faces atop a compressed-yet-malleable body? Yes, it appears so. Robot Dog Astro Can Sit, Lie Down, and Save LivesMIT’s AI Knitting System Designs, Creates Woven Garments Here are some AI generated nude portraits I’ve been working onUsually the machine just paints people as blobs of flesh with tendrils and limbs randomly growing out – I think it’s really surreal. I wonder if that’s how machines see us… pic.twitter.com/tYgzCHGfse— Robbie Barrat (@DrBeef_) March 27, 2018Among his other AI adventures, Barrat taught a neural network to rap (and subsequently caught the attention of NVIDIA, where he spent ten months as a deep-learning software intern) and turn Minecraft into a “playable oil painting.”My personal favorite undertaking is his artificial intelligence-generated landscapes: Quite stunning images that look like something worthy of a museum wall (or at least an Etsy shop).“One can tell that whatever painted these has little idea of how the physical world works,” Barrat wrote on his website, describing the “fairly realistic [but] strangely surreal” results as featuring “trees with floating branches, and multiple trees having joined trunks.”According to his LinkedIn profile, Barrat now works as a researcher at Stanford University School of Medicine, assisting in research “at the intersection of artificial intelligence and genomics.” Let us know what you like about Geek by taking our survey.