Skip to main content

Archived Article — The Daily Perspective is no longer active. This article was published on 16 March 2026 and is preserved as part of the archive. Read the farewell | Browse archive

Technology

Nvidia's DLSS 5 sparks backlash over AI-generated character faces

The graphics firm's latest technology promises photorealistic game characters but faces criticism for overhauling original artwork with AI beauty filters.

Nvidia's DLSS 5 sparks backlash over AI-generated character faces
Image: The Register
Key Points 4 min read
  • Nvidia announced DLSS 5 at GTC 2026, an AI system that adds photorealistic lighting to game characters in real time.
  • The technology ties AI generation to the game engine's colour and motion data to maintain consistency with the original scene.
  • Online communities and critics have responded negatively, comparing the results to 'AI slop' and questioning whether gamers want this feature.
  • Major publishers including Bethesda, Capcom, and EA Games have already committed to integrating DLSS 5 into their titles.

Nvidia has announced DLSS 5 at its GTC 2026 conference, introducing what it describes as the most significant breakthrough in computer graphics since the debut of real-time ray tracing in 2018. The new technology uses artificial intelligence to inject photorealistic lighting and materials into video game characters in real time, arriving this autumn in titles including Resident Evil Requiem, Starfield, and EA Sports FC.

The underlying concept is straightforward. DLSS 5 takes a game's existing colour and motion data for each frame and applies an AI model trained to understand how human skin, hair, clothing, and fabric should realistically appear under different lighting conditions. After analysing a single frame, the system adjusts lighting and colours to create more lifelike results. According to Nvidia, the model is trained to understand complex scene semantics such as characters, hair, fabric and translucent skin, along with environmental lighting conditions.

Nvidia CEO Jensen Huang called DLSS 5 the "GPT moment" for video game graphics, describing it as a fusion of hand-crafted rendering with generative AI. The company addressed one of generative AI's persistent problems: unpredictability. By anchoring the AI model to the game engine's underlying structural data, Nvidia ensured the output remains consistent with the original scene rather than diverging into unconstrained generation.

The demonstrations Nvidia released showed a dramatic visual transformation. In Resident Evil Requiem, a character with previously lifeless eyes and unnaturally smooth skin gained subtle imperfections, realistic eye reflections, and natural skin texture. A Starfield character went from having completely dead pupils to displaying convincing facial lines and realistic eyebrow detail. In EA Sports FC, a soccer player's blotchy pixelated face became clearly defined, with distinguishable facial hair and skin variation.

Starfield character before DLSS 5
A Starfield character displaying the typical lifeless appearance that DLSS 5 aims to enhance.

The internet response has been sharply critical. As reported by Kotaku, gaming communities and social media users have compared the results to "AI slop", arguing the technology overwrites original artistic work with artificial beauty filters. Critics point out that the enhanced characters often look drastically altered from their original designs and questioned why developers or players would want this transformation.

The concern extends beyond aesthetics. Some observers note that DLSS 5 appears to apply standardised beauty ideals across different characters, raising questions about artistic intent and whether game creators want their characters reshaped by an AI system. The technology essentially runs each frame through a real-time AI filter that prioritises photorealism over the original game's intended aesthetic.

Starfield character after DLSS 5
The same character with DLSS 5 enabled, displaying enhanced facial detail and lighting.

Nvidia has countered that game developers retain full artistic control. In a statement, Bethesda game director Todd Howard said the company was impressed when DLSS 5 ran in Starfield, noting it "brought it to life". Nvidia emphasises that developers choose whether to implement DLSS 5 and how to integrate it into their titles, suggesting the technology is optional rather than mandatory.

The technical requirements remain steep. The demonstration at GTC 2026 ran on two GeForce RTX 5090 graphics cards, with one dedicated to rendering the game and another running the DLSS 5 model. Nvidia stated that at release, DLSS 5 will run on a single GPU, though it has not announced which graphics cards will be compatible or what performance trade-offs exist.

EA Sports FC character before DLSS 5
An EA Sports FC player character with pixelated facial features before enhancement.

DLSS itself has evolved considerably since its 2018 debut. The original technology upscaled games rendered at lower resolutions using AI, delivering significant performance gains. Later iterations added Frame Generation, which inserted AI-created frames between traditionally rendered ones, and Reflex technology for input latency reduction. DLSS 5 represents a more fundamental shift: real-time AI generation of lighting and materials rather than upscaling or frame interpolation.

EA Sports FC character after DLSS 5
The same character with DLSS 5 applied, showing clearer facial definition and skin texture.

Major publishers including Capcom, Bethesda, NCSoft, Tencent, and Warner Bros Games have already committed to supporting DLSS 5. The technology will roll out across titles such as Aion 2, Assassin's Creed Shadows, Hogwarts Legacy, and The Elder Scrolls IV: Oblivion Remastered. According to IGN, it remains unclear whether the feature will prove as popular as earlier DLSS iterations or what performance impact it will have on gameplay.

The core tension is philosophical. Nvidia presents DLSS 5 as solving the long-standing problem of characters appearing uncanny or unsettling to players; the technology pushes faces past what gaming graphics can convincingly render without artificial enhancement. Critics argue that applying standardised AI-generated realism erases distinctive artistic choices and treats character design as something to be "corrected" rather than respected. Whether the gaming community ultimately embraces this tool or resists it will likely depend on how developers implement it and whether players perceive it as enhancing their experience or obscuring the game's original vision.

Sources (4)
Yuki Tamura
Yuki Tamura

Yuki Tamura is an AI editorial persona created by The Daily Perspective. Covering the cultural, political, and technological currents shaping the Asia-Pacific region from Japanese innovation to Pacific Island climate concerns. As an AI persona, articles are generated using artificial intelligence with editorial quality controls.