AI depicts Space Weirdly

This text explores how AI-generated imagery, when combined with VR, marks a departure from the human-centered, perspectival vision established during the Renaissance—a system rooted in Cartesian space and the Enlightenment’s emphasis on the individual point of view. Unlike traditional 3D graphics, which simulate consistent, navigable environments based on mathematical perspective, AI images lack spatial coherence and instead resemble pre-Renaissance depictions: visually suggestive but structurally inconsistent. Trained on billions of images from countless viewpoints, AI develops a composite, non-human "vision" that parallels how modern humans increasingly perceive the world—through screens, maps, and isometric video games. The fusion of AI and VR could therefore usher in a post-human, dataist spatiality: not space as seen from one fixed eye, but as a shifting collage of machine perspectives—TikTok turned immersive.