I don't think it's a stretch to imagine an AI creating "beauty." Even at today's state of the art I'm sure a properly configured AI could produce digitally or even, with some fine-grained robotic work, with paint on canvas, an original painting in, say, the impressionist style that would be regarded as beautiful by most who saw it. Would the AI itself experience the work as "beauty?" Today, not by a long shot, although I'm guessing that it would be fairly easy to equip "Watson" to evaluate the relative appeal to the Teeming Millions of a canvas by Francis Bacon and one by, say, Monet. That's obviously a long way from wiring a genuine aesthetic sensibility into a server farm.
I'm reasonably confident that whatever prodigies of electronic cognition we may see in the coming years, however, it is likely that our electronic cobblers will stick to their lasts, and that a dedicated community of Japanese claims adjustment algorithms will not leave off calculating benefits to compose haiku or to arrange rock gardens.
cordially,
I'm reasonably confident that whatever prodigies of electronic cognition we may see in the coming years, however, it is likely that our electronic cobblers will stick to their lasts, and that a dedicated community of Japanese claims adjustment algorithms will not leave off calculating benefits to compose haiku or to arrange rock gardens.
cordially,