"Assume every game company is using AI." Nexon's CEO says the edge now belongs to human creativity
Following the launch of ARC Raiders, Nexon CEO Junghun Lee shared a blunt assessment of where game development stands: AI is already embedded across production and live ops, and it's lifting the industry's baseline. That doesn't guarantee winners. It just removes excuses.
Context: AI is everywhere, and players notice
ARC Raiders - like The Finals before it - drew scrutiny for using AI tools, especially text-to-speech for certain voice lines. The developer has said its visuals don't rely on generative AI and has defended using automation to keep team size sustainable. The signal underneath the noise: modern pipelines now mix human talent, machine learning, and strict constraints to ship faster without ballooning headcount.
Lee's core point: efficiency is table stakes; creativity is the moat
"AI has definitely improved efficiency in both game production and live service operations," Lee said, adding that the industry's "average" quality has been rising. His follow-up carries the real weight: "It's important to assume that every game company is now using AI. But if everyone is working with the same or similar technologies, the real question becomes: how do you survive?"
His answer: "human creativity." If AI raises the floor, taste, judgment, and original direction decide who breaks out.
What this means for executives, product leaders, and engineering heads
- Strategy: Decide where AI is commodity and where it's differentiation. Codify 2-3 creative pillars that no model can replace (tone, world-building, systems feel).
- Org design: Keep teams lean but senior. Fewer, stronger leads who can set vision and say no. Spin up an AI enablement group that builds internal tools and guardrails.
- Production: Standardize data pipelines, asset provenance, and rights management. For TTS or synthetic media, require talent consent, clear labeling, and human review.
- Live ops: Use models for triage, moderation, and dynamic tuning, but keep humans in the loop for balance and narrative decisions.
- Governance: Publish a public-facing AI policy. Disclose AI usage where it touches the player experience. Align with legal on IP, likeness, and union agreements.
- Trust: Communicate early on store pages and patch notes. Explain what's automated and why. Transparency beats speculation.
- Metrics that matter: Cycle time per asset, defects found pre-release, content freshness in live ops, sentiment on AI usage, and-most important-distinctiveness players can feel.
- Risk management: Don't let models flatten your game's voice. Guard against sameness. Audit for bias, compliance, and vendor lock-in.
Practical next steps (90-day window)
- Run a portfolio-wide audit: where AI is used today, what it saves, and where it may harm creative identity.
- Define a "human-in-the-loop" standard for writing, voice, art style, and economy design.
- Pilot one internal tool that removes a bottleneck (e.g., automated asset tagging or QA triage) and one creative tool that's sandboxed with senior oversight.
- Stand up an internal style bible so AI outputs don't drift from your studio's signature.
- Set disclosure rules for store pages and community posts; make them consistent across titles.
Bottom line: assume AI gives everyone speed and scale. Your advantage is the taste, constraints, and editorial judgment that models don't have. Build for that, and let AI do the busywork.
If you're building team-wide AI fluency and governance, here's a practical starting point for role-based upskilling: AI courses by job.
Your membership also unlocks:
