‘Blue Archive’ Director Says Audiences Are Tired Of “AI Slop”, Want “Creative Authenticity” From Their Video Games
With the frequency of generative AI use in video game development rising just as rapidly as players’ rejection of it, Blue Archive director Yongha Kim has attempted to explain how the issue is not with the technology itself but its application, as “slop” creators fail to recognize their audiences’ desire for “creative authenticity”.
Asked for his thoughts on the growing “resentment toward AI-generated content, like illustrations” during a recent interview with South Korean video game news outlet GameMeca, Kim asserted, as machine translated by DeepL, “I empathize with gamers’ concerns and believe it naturally impacts (content acceptance, etc.)” before proceeding to “address this from two perspectives.”
“First, there’s the issue of AI slop—where indiscriminate use of generative AI lowers the quality of the output. To use a snack food analogy: if the packaging looks impressive but the actual snack content is reduced and replaced with nitrogen, consumers will naturally react negatively.
“Second, consumers of subculture genres have high expectations for the ‘authenticity of creativity’. Conversely, current transformer or diffusion-based models are merely simulators, not entities with intent or personality. If we rely entirely on them to churn out results with a mere ‘click,’ it raises the question: ‘Can such an approach truly embody the creator’s authenticity?’
From there pressed as to whether he believed AI-generated content could “completely replace human-created works in the near future”, Yong, who also serves as the CEO of Blue Archive‘s publisher Nexon, defined ‘near future’ as “next year” and argued, “Speaking from this perspective, my answer is ‘no’. The core issue is that AI still cannot fully generate the level of output demanded in development environments.”
“I anticipate it will still fall short of meeting expectations next year. However, it is undeniable that its utility as a tool is expanding.”
Turning to the limitations of current AI technology, Kim added, “From a technical perspective, limitations include the lack of continuous learning capabilities, constraints on the amount of contextual input, performance limitations in modalities beyond text (such as images, voice, video, etc.), and the enormous costs involved in training and utilization.
“From a creative standpoint, it still lacks the ability to independently propose novel ideas—things that are genuinely ‘new and unprecedented’—that humans would prefer. This is because current generative AI research prioritizes producing outputs that align with verifiable correct answers and public consensus.
“Additionally, while it can answer like an expert in specific domains, it often provides surprisingly poor answers in common-sense areas or generates hallucinations, lacking the consistency humans possess. Therefore, rather than a blanket approach like ‘Let’s all use AI for work!’, it is necessary to set the scope of application and utilization methods in a localized and concrete manner.”
Notably, this is not the first time Kim has weighed in on the topic of generative AI in video games.
Giving a lecture on ‘Developing Cute Girl Games in the AI Era‘ to attendees of South Korea’s 2025 Content Creative Talent Mentoring Program, Kim similarly explained, “Users who enjoy character games do not prefer AI-generated content. I also feel a strong aversion to spending money on images that are simply ‘clicked’ into existence.”
“We need to approach this from the perspective of how we can improve the convenience of those who are currently developing. I think that a position that provides focused support for such work is extremely important. As such a position, I believe that the roles of machine learning engineers and technical artists will become increasingly important in game development.”
