“Well, first of all, they’re completely wrong,” Huang said in response to a question from Tom’s Hardware editor-in-chief Paul Alcorn about the criticism.
“The reason for that is because, as I have explained very carefully, DLSS 5 fuses controllability of the of geometry and textures and everything about the game with generative AI,” Huang continued.
Just a elongated way to say AI slop.
From what Nvidia’s showing (and what Huang emphasized in that Tom’s Hardware Q&A at GTC 2026), DLSS 5 is pitched as “neural rendering” or “content-control generative AI.” It uses the game’s existing 3D data (motion vectors, scene semantics like hair/skin/fabric, lighting conditions) as grounded input, then the AI infuses photoreal lighting, materials, and enhancements while staying consistent frame-to-frame. Devs get tools to fine-tune intensity, color grading, masking, etc., so they can dial in how much “photorealism” gets applied without losing their artistic intent
Wow, that’s an even longer way, even more so than the CEO defending his AI Slop, actually slightly impressive in a corporate sub way, to say AI Slop.