Keywords Studios, one of the biggest third-party service providers in the games industry, has evaluated more than 500 AI tools for game development use - and the verdict is sobering. According to the company's head of transformation, only about half a dozen of those tools actually help "in the right way," as reported by Video Games Chronicle.
The findings paint a picture of an AI landscape that is long on hype and short on practical value. Keywords Studios works with a huge range of developers and publishers across art, audio, localization, and QA, which gives the company a uniquely broad vantage point when assessing which tools actually move the needle in a real production environment.
Signal vs. noise
The head of transformation described the current moment bluntly: "AI feels like it's in the chaos phase right now." That framing resonates with what many developers have been saying quietly for a while - the sheer volume of AI products hitting the market makes it genuinely difficult to identify tools that solve real problems versus those that demo well and deliver little.
Out of 500-plus tools evaluated, landing on roughly six that clear the bar is a conversion rate that should give the industry pause. It also raises questions about how much time and money studios are burning just trying to assess what is actually useful before a single line of code gets written or a single asset gets shipped.
What this means for the broader industry
Keywords Studios' scale matters here. The company's evaluation isn't the opinion of a single studio with one genre and one pipeline - it's a stress test across dozens of workflows, disciplines, and production contexts. When a company with that kind of coverage says the useful tools number in the single digits, that carries real weight.
That said, "chaos phase" doesn't mean "dead end." Keywords' framing suggests this is a transitional moment rather than a verdict on AI in games full stop. The useful tools do exist - they're just buried under a mountain of noise that requires significant resources to sift through, resources that smaller studios simply don't have.
For developers trying to navigate AI adoption right now, the takeaway is essentially to be skeptical, move deliberately, and lean on evaluations from organizations that have already done the heavy lifting. The gold rush mentality around AI tooling may be doing the industry more harm than good at this stage.
