In an attempt to teach AI systems to design book covers independently, data scientists at the Kyushu University in Japan have created an algorithm that teaches computers to determine book genres by their cover.


Data scientists at Kyushu, Brian Kenji Iwana and Seiichi Uchida, created a program that would allow computers to compare book covers with their content to judge what sorts of designs people associate with certain genres. Based off of a sample size of 137,788 book covers across 20 genres, “the algorithm listed the correct genre in its top 3 choices over 40 percent of the time and found the exact genre more than 20 percent of the time,” doing the best at determining cookbooks and travel books. Despite the need for more experimenting, the idea of an AI either providing aid for book designers or completely taking over this task leaves a lot of room for speculation on the the future of book designers.