What happens when artificial intelligence, trained on a diet of vintage erotica and PornHub clips, is let loose to generate its own images? For digital artist Eric Drass, the result is a hallucinatory cascade of melting flesh, twisted limbs, and haunting faces that defy easy description. His ongoing project, "The Machine Gaze," uses machine learning tools such as Nvidia's StyleGAN and Google's DeepDream to produce a body of work that is as provocative as it is disorienting.
Drass, who works from his studio in Brighton on England's south coast, has spent years training models on a vast array of imagery sourced from vintage erotica and PornHub. The output, he explains, is not pornography but something else entirely—a fleshy, abstract echo of its source material. "By turning the AI on the domain of porn you make imagery that's not pornography but it's fleshy—you can see the roots that it's come from," Drass told Futurism. "It starts to trigger the similar parts of your brain but in slightly the wrong way."
The resulting videos, which Drass pairs with his own musical compositions, veer between titillating and nauseating. Bodies dissolve and fuse into one another in a relentless, morphing stream that can feel like a fever dream. His still images, meanwhile, capture faces at their most liminal—awful masks of madness that seem to form as the figures warp between states. The overall effect is to make you question what you are looking at, and why it feels so unsettling.
What Machines See in Porn
At the core of the project is a philosophical inquiry: what does a machine, with no concept of sexuality or desire, actually see when it looks at pornographic content? Drass notes that a huge proportion of internet traffic is porn, yet it remains a largely unspoken part of our digital lives. "But what about the machines that are observing our data, observing our behavior?" he asked. "To a machine watching the internet, it will see 30 percent of it is these monkeys looking at other naked monkeys. It's fucking crazy!"
His work, he says, is an attempt to turn that algorithmic gaze back on itself, forcing us to confront how AI perceives human intimacy. The ambiguity of the images—are they hot, grotesque, or both?—is central to the project's impact.
Deepfakes and Algorithmic Censorship
Drass is also keen to distance his work from the phenomenon of deepfakes, which often use similar technology to superimpose faces onto pornographic videos without consent. He calls such uses "morally objectionable" and a "startling lack of imagination." Instead, he sees his own work as a creative exploration of what AI can reveal about our own desires.
The project has also sparked an interesting side experiment in algorithmic censorship. Drass posted a promo video for "The Machine Gaze" on various social platforms to see if their algorithms would flag it. LinkedIn was the only platform that didn't block it; Facebook, Twitter, and YouTube all shut it down. "There is a weird tension, because this stuff is not naked people, just what machines are judging to be naked humans," he said. "So who deems it filthy? It's just a picture."
As for whether machines will ever experience pleasure themselves, Drass is skeptical. "Pleasure is about being in a body," he said. "The fact that we're made out of biology and have a direct relationship with the world, machines cannot currently grasp."