Gamers are "completely wrong" about DLSS 5-Nvidia's CEO has responded to criticism of the new technology.

“DLSS 5 combines control over geometry, textures and everything else in a game with generative AI.”

At a GTC 2026 press conference hosted by Tom’s Hardware, Nvidia CEO Jensen Huang addressed criticism of DLSS 5, the company’s new technology built around artificial intelligence and neural rendering.

Since the feature was announced, some critics have complained on social media that the technology reduces visual quality in games, makes them look more uniform, or reflects only Nvidia’s perspective. Much of that criticism focused on the updated appearance of Grace Ashcroft and Leon Kennedy in Resident Evil Requiem.

Nvidia, Jensen Huang and DLSS 5: what he says developers control

“Well, first of all, they’re completely wrong,” Huang said when Tom’s Hardware editor-in-chief Paul Alcorn asked about the backlash. “The reason is that, as I explained in great detail, DLSS 5 combines control over geometry, textures and everything else in the game with generative AI.”

He added that developers can still “fine-tune the generative AI” to match their own style, stressing that DLSS 5 introduces generation capabilities on top of a game’s existing geometry, but that this “does not change artistic control”.

“This isn’t post-processing, it isn’t frame-level post-processing, it’s generative control at the geometry level,” he said.

Huang also said studios can try the tool and decide how they want to apply it, suggesting it is ultimately up to the developer whether to create a “cartoon shader” or make the game look different.

“All of this is under the control - direct control - of the game developer,” he said. “That’s very different from generative AI; it’s generative AI with content control. That’s why we call it neural rendering.”

DLSS 5 is scheduled to launch in autumn.