Machado and Cardoso’s NEvAr system

Machado and Cardoso’s NEvAr system is an artistic tool that leverages concepts from evolutionary art. It is thus a continuation of decades of evolutionary and genetic algorithm art, which shares machine learning roots with today’s AI art. Developed by Penousal Machado and Amílcar Cardoso at the University of Coimbra, the NEvAr (Neuro Evolutionary Art) system harnesses evolutionary algorithms to automate aesthetic decision-making. Users engage with NEvAr by evaluating and selecting images, which then undergo evolutionary processes to enhance their appeal based on the users' preferences. This system utilizes a fitness function that incorporates complexity measures, aligning with Philip Galanter’s exploration of evolutionary art’s potential to overcome the “fitness bottleneck” in creative processes. NEvAr’s capacity to generate visually compelling artwork with limited human intervention underscores how evolutionary artists were pushing for—or creating their own—tools that today resemble more automated processes like GANs and generative AI. Like many evolutionary art and early AI work, NEvAr was inspired by the work of Karl Sims as well as Richard Dawkins. Yet its roots extend to Yoichiro Kawaguchi from Japan.