Remember when loot boxes were going to revolutionise monetisation? Or when the metaverse was six months away from changing everything? The games industry has a well-documented habit of collectively losing its mind over new technology, and AI is the latest casualty of that pattern.
Except this time, the stakes are higher, and the budgets are bigger.
Jon Gibson, Head of Transformation, put a number on it recently that nobody in the industry seems to want to discuss too loudly. His team at Keywords Studios reviewed around 500 AI tools. The ones that held up in actual production? A small fraction of that total.
Five hundred tools evaluated. A handful worth using. That’s not a technology problem. That’s an industry that has confused being busy with making progress.
What Happens When the Applause Dies Down
Here’s something vendors will never put in a pitch deck: their tool has never met a real game in production.
It hasn’t dealt with the codebase that three different studios touched over seven years. It hasn’t navigated an art pipeline held together with custom scripts and institutional memory. It hasn’t been handed to a team of exhausted developers two months from ship who have neither the time nor the appetite to learn a new system from scratch.
Demos are frictionless by design. Production is friction by nature. And that chasm is where the industry’s AI ambitions keep falling quietly to their death.
Gibson put his finger on it precisely: “Everyone’s focusing on building better AI, and no-one’s really focusing on how to use it in a live production environment.”
The uncomfortable follow-up question nobody is asking loudly enough is: why? Why has an industry full of technically sophisticated people continued to evaluate tools against demo conditions rather than production realities? Partly hype, partly pressure, partly the very human tendency to want the exciting answer to be the right one.
Backwards by Design
The most damaging thing about how studios are currently approaching AI isn’t the money being wasted, though that’s real. It’s the logic being applied.
The standard playbook goes something like this: hear about a tool, get excited about the tool, find a reason to use the tool, report back that AI exploration is underway. Somewhere in that sequence, the actual question of what problem needed solving never quite gets asked.
Gibson is direct about this: “A lot of people focus on what’s cool. They focus on the tool itself or the model itself, rather than what they’re trying to do.”
“A company will use a tool or build a tool without a specific use case and try and cram it into their production pipelines, rather than flipping that problem around and saying: ‘What are our pain points? What are we trying to solve?’ And then building a tool against that.”
It sounds obvious when it’s laid out that plainly. And yet the industry keeps doing it, because starting with a problem is slower and less exciting than starting with a tool that generates impressive-looking outputs.
The People Being Ignored
There’s a conversation happening in boardrooms and strategy sessions about AI adoption. There’s a different, quieter conversation happening among the developers who would actually live with the consequences of those decisions.
Those two conversations are not connected.
Gibson noted something that should reframe the entire adoption debate: “That statistic of 52% of developers being concerned about the usage of AI, that’s gone up every year for the last three years. As AI tools and AI models and AI technology has become more prevalent, the lack of understanding and the concern has increased.”
Think about what that trajectory actually means. This isn’t a fear-of-the-new problem that familiarity will eventually dissolve. Developer concern is growing in direct proportion to AI’s presence in the industry. The more real it becomes, the more worried people are.
The reasons aren’t mysterious. Game development is a creative discipline built on craft, authorship and collaboration. Questions about what gets automated, who owns the output, and what it means for careers aren’t abstract philosophical worries. They’re about people’s livelihoods and identities as makers. Studios that treat those concerns as a communications problem to be managed rather than a human problem to be taken seriously are going to struggle to implement anything that sticks.
Boring Is the New Breakthrough
The exciting version of AI in gaming gets written about constantly. Generative worlds. Infinite content. Procedural everything. It’s compelling stuff, and some of it will eventually be real.
The useful version of AI in gaming is much less glamorous. It’s a clearly defined bottleneck in a specific pipeline, matched to a tool that a specific team can actually use without disrupting everything around it, governed by sensible policies and measured against outcomes that someone has actually defined in advance.
Less TED talk. More spreadsheet.
Gibson describes the current moment as a chaos phase. Getting out of it doesn’t require better technology. It requires studios willing to do the unglamorous work of figuring out what they actually need before they go looking for something to buy.
The tools will still be there. The question is whether the thinking will finally catch up.
Discover more from The Game of Nerds
Subscribe to get the latest posts sent to your email.
