The Challenges of AI in Game Development
Jon Gibson, the Head of Transformation at Keywords Studios, has shared insights on the current state of AI tools in game development. According to Gibson, while many game studios are embracing AI-powered tools, only a small number of these tools are actually effective and address real-world problems.
Gibson spoke with The Game Business about his experience with testing around 500 AI game development tools. These tools were evaluated for their ability to streamline workflows and improve efficiency. Keywords Studios often plays a support role, providing additional development for major titles such as Batman: Arkham Shadow, Still Wakes the Deep, and Goat Simulator 3.
Despite this extensive testing, Gibson noted that there are only approximately half a dozen AI tools that truly complement development in the right way.
Why Most New AI Tools Aren’t Useful for Game Development
One of the main issues with most AI tools is that they lack a formalized purpose and aren’t built for practical, long-term use. Gibson explained that many developers are focused on creating better AI models rather than figuring out how to implement them effectively in live production environments.
“AI feels like it’s in the chaos phase right now. And we need to move to the usable phase,” Gibson said. “How do we use AI in live production environments? How do we use AI in a way where it complements teams rather than potentially threatens teams? And also how do we use it in a context where it’s governed, it’s controlled, it’s IP safe, it’s legally safe, it’s ethically and morally safe?”
Gibson highlighted a significant gap between the flashy AI demos shown to developers and the public and the practical implementation of these tools. He emphasized that achieving consistent, high-quality results requires human oversight and direction.
The Focus on What’s ‘Cool’ Rather Than Solving Problems
Gibson pointed out that many new AI tools focus on what is considered ‘cool’ on the surface, such as producing bright visuals without a solid coding foundation. This approach often overlooks the actual needs of developers.
“A lot of people focus on what’s cool instead of solving problems,” Gibson said. “They focus on the tool itself or the model itself, rather than what they’re trying to do.”
He added that companies often use or build tools without a specific use case, trying to force them into production pipelines. Instead, Gibson suggests flipping the problem around by identifying pain points and building tools to address them.
The Problem of ‘Garbage’ Tools
With over 500 tools tested for practical use at Keywords Studios, only a handful have proven to be both practical and useful. Gibson described the current landscape as filled with ‘garbage’ tools that don’t meet the necessary standards.
Another flaw in the current thinking about AI implementation in game development is the tendency for studios to throw everything at the wall. They introduce AI tools without explaining their purpose to staff or providing clarity about their intentions.
This lack of communication can lead to confusion and concern, particularly regarding job security. Gibson mentioned that 52% of developers are concerned about the use of AI, and this percentage has increased every year for the last three years.
The Need for Meaningful Integration
Until AI technology is meaningfully integrated into production pipelines in a safe, valuable, and complementary way, Gibson believes concerns about its impact on the industry will persist.
“I think there’s always going to be that level of concern,” he said. “Until we’re able to work out how to build this technology into production pipelines in a way that adds value, complements existing talent, and doesn’t threaten it, the concerns will continue.”
