Jensen Huang has shared a new explanation of DLSS 5, but we’re not sure what he’s trying to say.
In a Q&A following his GTC 2026 keynote, Huang said this:.
…It’s not post-processing, it’s not post-processing at the frame level, it’s generative control at the geometry level.
He also reiterated that developers have control over how much AI DLSSS 5 uses, saying:
All of that is in the control — direct control — of the game developer. This is very different than generative AI; it’s content-control generative AI. That’s why we call it neural rendering.
Nvidia previously stated that it DLSS 5 not applying an AI filter. These new statements seem to muddle the issue.
For what it’s worth, DLSS is an acronym for deep learning super sampling, a form of upscaling. What Nvidia revealed and Jensen describes here are different technologies from upscaling.
Beyond explaining this technology to the layman, Nvidia is struggling to get gamers to accept the idea. With the stigma that accompanies labels like generative AI, this hasn’t really assuaged consumers’ worries.