It has been two days since Nvidia unveiled DLSS 5, coming this Fall to “transform visual fidelity in games”. The fifth-gen tech does so with a new real-time neural rendering model that “infuses pixels with photoreal lighting and materials”. Initial reaction from gamers has been mixed, with many fearing it will hurt artists intent and basically make everything look like an AI filter. And now, Nvidia CEO Jensen Huang himself has come out to address the backlash.

Nvidia announced DLSS 5 at GTC 2026, which was quickly followed by debates, complaints, and plenty of memes online. In a later Q&A session during GTC, Huang explained why critics are “completely wrong” about DLSS 5. In particular, addressing the ‘AI filter’ comments that seem to suggest DLSS 5 is some kind of post-processing technique with a (digital) filter.

Clearing up misconceptions about DLSS 5

DLSS 5 aims to bring photorealistic visuals by transforming lighting and materials. Anyone with knowledge of studio photography will know how much different lighting can make to a scene, resulting in some pretty wild differences. In any case, during the Q&A attended by Tom’s Hardware, Huang was keen to point out that DLSS 5 “is not post-processing, it’s not post-processing at the frame level, it’s generative control at the geometry level”.

“Well, first of all, they’re completely wrong. The reason for that is because, as I have explained very carefully, DLSS 5 fuses controllability of the geometry and textures and everything about the game with generative AI.”

Jensen Huang, Nvidia CEO and co-founder

Huang also points out that developers can try and use DLSS 5’s neural rendering model to make a “toon shader” or transform visuals so it looks like everything is “made of glass” – so it seems it’s not exclusively about photorealism.

Backing up previous comments from Nvidia and devs like Bethesda, the CEO says, “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,” he continued. It’s still generative AI, but not the same method as asking ChatGPT to generate an image.

DLSS 5 is designed in tandem with developers, ultimately giving them artistic control, and the neural rendering model is designed to keep visuals consistent “from frame to frame”. Since DLSS 5 doesn’t launch until this Fall, we expect there will be more demos in the meantime, ideally with more variety, and we’re sure the internet will continue to have a lot to say.

About the Author

Jack Goodall

At PC Guide, Jack is mostly responsible for reporting on hardware deals. He also specializes in monitors, TVs, and headsets and can be found putting his findings together in a review or best-of guide.