Speaking to Japanese publication GameSpark, Nexon CEO Junghun Lee said that he thinks “it’s important to assume that every game company is now using AI.”
This was in the context of being asked about the impact of generative AI technology in regards to Nexon’s future projects—a tech which game developers are still not warming up to, and some of them are becoming increasingly worried that it will lower the quality of games.
Translated by Automaton, Lee said that “AI has definitely improved efficiency in both game production and live service operations,” leading to the personal consideration that the “average” quality level of games is on the rise thanks to it.
Yet, returning to the assumption that everybody is on board with the gen AI train, the CEO wonders how studios can stand out amid that competition.
“If everyone is working with the same or similar technologies, the real question becomes: how do you survive? I believe it’s important to choose a strategy that increases your competitiveness,” he said.
To Lee, the strategy is “human creativity.”
Nexon has been working on AI tools for eight years
While the CEO was talking about future Nexon projects, it’s unclear how or in which capacity AI will be implemented during the game development process. In 2017, the company launched Intelligence Labs, an in-house AI research and development team, which can help paint a picture.
Via The Korea Herald, Intelligence Labs has used machine learning and deep learning to “come up with more engaging and immersive games,” spearheaded by a team of around 700 engineers that is now “zeroing in on generative artificial intelligence research to create original, personalized content for players.”
In a more practical example, the team publicly launched Gamescale in April 2023, which offers a game matchmaking algorithm based on user preferences and playstyles, posing an alternative to skill-based matchmaking.
Other tools include the Nexon Voice Creator, a text-to-speech system that creates voices in different accents and pitches for nonplayer characters without separate recordings. The team has also been testing the “AI NPC” function, allowing NPCs to “directly communicate with users beyond a single script.” As noted by The Korea Herald, this opens the possibility of creating “false, biased, discriminatory, and abusive remarks.” To this, Intelligence Labs said it’s committed to “putting effort into building ethical guidelines for trustworthy AI in gaming.”
Lee’s comments on the use of AI at the company took place amid a wave of criticism around Arc Raiders, the extraction shooter developed by Embark Studios and published by Nexon. Over at the game’s Steam page, the studio has an AI-generated content disclosure, which states the following: “During the development process, we may use procedural-and AI-based tools to assist with content creation. In all such cases, the final product reflects the creativity and expression of our own development team.”
Speaking to PCGamesN, Arc Raiders design director Virgil Watkins said the game “in no way uses generative AI whatsoever,” adding that there is “something called machine learning, or reinforcement learning, and that’s to do with the locomotion for our larger drones with multiple legs,” but remarked that there’s “no generative content whatsoever.”
That being said, the text-to-speech system present in The Finals, also from Embark, which is a speech tool that generates audio voice lines for in-game commentators, is used in Arc Raiders as well.
“As [you] stated, [it’s] the same as The Finals, we use that text-to-speech model. That is, we hire and contract voice actors for it—it’s part of their contract that we use it [AI] for this purpose, and that allows us to do things like our ping system, where it’s capable of saying every single item name, every single location name, and compass directions. That’s how we can get that without needing to have someone come in every time we create a new item for the game,” Watkins told PCGamesN.
In an interview with Eurogamer, Embark CCO Stefan Strandberg said the studio uses “AI as tools to assist in some content creation, but it’s always in the context of creativity, as an expression of the team.”
Elaborating on the text-to-speech model specifically, Strandberg said that the tool allows the team to “increase the scope of the game in some areas” where Embark thinks it’s needed, or where there’s “tedious repetition” in situations where voice actors “may not see it as valuable work.”
The CCO’s justification, then, is that these tools are allowing smaller teams to scale their operations. “We started this studio with a massive set of procedural tools when it comes to world building, and we wouldn’t be anywhere near where we are today without those investments,” he said.”
