The developer of an AI-powered driving instruction simulator is spending $1,000 per month just to keep its demo running
The article discusses the financial challenges of implementing AI-powered NPCs in games, using “Teach My Little Sister How To Drive” as a case study. This upcoming driving instruction simulator by FRVR features an LLM chatbot that powers the virtual sister character players coach on driving technique from the passenger seat. However, the developer is spending $1,000 per month just to keep the game’s demo running, illustrating why relatively few games have adopted this technology despite its potential.
While AI-powered chatbots could theoretically create more reactive and intelligent non-player characters, practical implementation has proven challenging. Games like “Where Winds Meet” have attempted to incorporate this technology, with mixed results so far. The substantial operational costs of running LLM services appear to be a significant barrier to broader adoption across the industry. This economic constraint may explain why LLM-powered NPCs remain uncommon in gaming, despite considerable industry hype about AI’s potential to revolutionize game development. The high cost per interaction suggests that until LLM technology becomes more efficient or less expensive, developers will likely continue to rely on traditional AI systems for NPC behavior.