Why many people dislike NVIDIA DLSS 5 (but will come to appreciate it in the future)

      Upscaling, or the real-time reconstruction of frames for video games, is a highly debated practice. Purists disapprove, while users with less powerful or mid-range gaming systems value the smoother performance it provides. NVIDIA engages in this, as do AMD and Intel. However, chaos ensued when Nvidia revealed the latest version of its super-sampling technology, especially due to the excessively AI-influenced appearance of the visuals, particularly in human faces.

      DLSS 5 Off (left) vs DLSS 5 On (right) NVIDIA

      The past few weeks in the tech industry have been quite eventful. If you have been following the DLSS 5 (Deep Learning Super Sampling) narrative, you’re likely familiar with the whirlwind of reactions ranging from “Wow” to “Wait, what?” and “Keep that away from my game.” Here’s a summary of the DLSS 5 controversy, starting with all the hype and leading to the current state described as a “2D filter.”

      The Story So Far: The “GPT Moment” That Wasn’t

      It all began when Jensen Huang took to the stage at NVIDIA’s GTC 2026 and made a shocking announcement: DLSS 5. NVIDIA was not merely upscaling pixels anymore; they claimed to be generatively reimagining them. Jensen referred to it as the “GPT moment for graphics,” assuring that AI would now take on the significant task of enhancing visual realism, including elements like skin texture, fabric shine, and intricate lighting. Sadly, the excitement didn’t even last for a full day.

      Awful actually😑 The more you look at it the worse it gets. This is basically Nvidia DLSS 5 applying its AI photorealism onto existing games. This is stripping the artistry from the game. For example: Resident Evil Requiem from Digital Foundry pic.twitter.com/qgoK4itztF— SpawnYaard 🎮 (@SpawnYaardReply) March 16, 2026

      Within hours, numerous side-by-side comparisons of Resident Evil Requiem and Starfield emerged online, and the community’s reaction? "AI Slop." Instead of enhancing the games, DLSS 5 was “Yassifying” characters by smoothing out rough skin textures, adding unintended makeup, and making everyone resemble an Instagram influencer from 2022.

      So … it turns out the studios that “created” those DLSS 5 clips weren’t informed about it either. It was clearly NVIDIA, which is why they all appeared similar. pic.twitter.com/Ubg3iECF6p— Beyond FPS (@beyond_fps) March 18, 2026

      Then came the “Betrayal.” According to reports from Insider Gaming, significant game developers were caught off guard. Artists at Ubisoft and Capcom learned about the DLSS 5 demos concurrently with the rest of us. NVIDIA quickly moved to mitigate the situation, promising a “Full Creative Control” SDK with adjustable intensity sliders. However, it became apparent just days ago from an email interview between YouTuber Daniel Owen and NVIDIA’s Jacob Freeman that DLSS 5 was not actually utilizing the deep 3D geometry of games. Essentially, it is a high-end 2D post-processing filter applied over the visuals. The “Neural Revolution” turned out to be merely a costly layer of paint.

      Why “Better” Isn’t Always Better

      On the surface, DLSS 5 seems impressive. And in certain contexts, it is. In instances like landscapes or static environments, the AI-enhanced shadows and highlights appear objectively “cleaner.” But the issue is that cleaner doesn’t always align with the intended aesthetic.

      Video games are an art form, and art encompasses intention.

      When a developer dedicates three years to perfecting a dim, atmospheric, claustrophobic hallway in a horror game, they don’t want an AI coming in to “improve” it.

      NVIDIA

      DLSS 5 tends to brighten dark areas and eliminate atmospheric fog because it perceives these as “errors” needing correction. The fact that developers were surprised by the demo is a significant warning sign. It reflects the typical corporate hierarchy: executives at the top approve things for marketing’s sake, leaving the creative teams unaware. Had NVIDIA collaborated with the artists from the beginning, it could have given the AI access to real 3D data models and blueprints.

      NVIDIA

      Imagine if the AI understood precisely where a character’s scar was meant to be or how a particular fabric was supposed to reflect light. In fact, recent examples showcased by Veedrac on Reddit revealed that games employing DLSS 5 with tone-mapping can look spectacular. This indicated that the technology can be effective, but only when a human directs its application. By launching it as a “black box” filter, NVIDIA effectively sidelined the very individuals who create engaging games.

      Additionally, there’s the significant issue of Data Sovere