Back in March, Nvidia introduced DLSS5, the newest version of its upscaling and frame generation technology. This launch quickly drew attention, though not for the reasons Nvidia likely anticipated. The technology faced criticism for its visual output, especially in the game Resident Evil Requiem, where it altered the character Grace Ashcroft into what some referred to as “yassified Instagram models.” This significant change triggered a wave of negative feedback from both developers and gamers.
In spite of the criticism, Nvidia CEO Jensen Huang defended DLSS5, claiming that detractors misinterpreted the technology and its potential. He maintained that DLSS5 was fundamentally distinct from generative AI, although this did little to shift public perception.
Amid the controversy, Masato Kumzawa, the producer of Resident Evil Requiem, presented a more hopeful viewpoint. He indicated that the uproar regarding DLSS5’s effect on character design demonstrated that the original design was robust and memorable enough to elicit such a response. Kumzawa’s remarks suggest a silver lining, indicating that the game’s core artistic vision resonated with audiences, even if the execution through DLSS5 was imperfect.
As the discussion persists, both the gaming community and developers are left to reflect on the implications of such technology for game art and design. While DLSS5 may have faltered in its initial reception, it also underscores the ongoing dialogue about the role of AI and upscaling technologies in the future of gaming.