According to Wccftech, Square Enix is making a major push into AI for game development with plans to automate 70% of QA and debugging tasks by the end of 2027. The Japanese publisher revealed this ambitious target in its latest earnings report as part of its Medium-Term Business Plan initiatives. They’ve already launched joint research with the Matsu Laboratory at the University of Tokyo specifically aimed at improving game development efficiency using AI technology. The current research team includes more than 10 people from both the Matsu-Iwasawa Laboratory and Square Enix engineers. This automation push is intended to “improve the efficiency of QA operations and establish a competitive advantage in game development.” The announcement comes as other major publishers like EA have already implemented AI in testing workflows, leading to significant layoffs at companies like Respawn where over 100 QA staff were cut last spring due to AI adoption.
The human cost of automation
Here’s the thing – when companies talk about “efficiency improvements” in QA, what they really mean is cutting costs and potentially cutting jobs. We’ve already seen the proof with EA’s moves at Respawn. Over 100 people lost their jobs because AI started handling feedback summarization from play testers. That’s work that used to require human judgment and context. Now Square Enix wants to take this even further by automating 70% of the entire QA process. That’s not just trimming around the edges – that’s fundamentally reshaping how game testing works.
And let’s be real – game testing has never been the most glamorous job in the industry, but it’s been a crucial entry point for many people wanting to break into games. If AI takes over 70% of these roles by 2027, what happens to that pipeline? Do we just accept that fewer people will get their foot in the door? The timing here is pretty aggressive too – we’re talking about three years to completely transform how a major publisher handles quality assurance. That’s lightning fast in corporate terms.
This isn’t just a Square Enix thing
Look, Square Enix might be making headlines with their specific 70% target, but they’re definitely not alone. High-profile Japanese developers like Masahiro Sakurai and Hideo Kojima have both expressed openness to using AI for tedious tasks. Their argument makes sense on paper – let AI handle the boring, repetitive work so humans can focus on creativity. But where exactly do we draw the line between “tedious” and “essential human judgment”?
The competitive pressure here is intense. When one major publisher starts achieving significant cost savings through AI automation, others basically have to follow or risk being at a disadvantage. We’re looking at a potential industry-wide shift where the first movers set the new normal. And with development costs for AAA games routinely hitting hundreds of millions, the financial incentive to automate is enormous.
But what about game quality?
Here’s my big concern – will AI-driven testing actually catch the weird, unexpected bugs that make games memorable (for all the wrong reasons)? Human testers bring context, intuition, and sometimes just random luck to finding issues that strict automated testing might miss. I’ve lost count of how many game-breaking bugs have slipped through even with extensive human testing – what happens when we remove most of those humans from the equation?
The other question is whether this actually speeds up development in meaningful ways. Sure, automated testing can run 24/7 and catch regression issues quickly. But debugging complex, interconnected systems often requires human understanding of how different game systems interact. Can AI really replicate that level of systemic thinking by 2027? That seems optimistic at best.
Square Enix’s full progress report outlines their broader AI strategy, but the QA automation target is definitely the most concrete and immediate application they’re pursuing. This feels like the beginning of a much larger transformation across the entire games industry – one that’s likely to reshape not just how games are tested, but who gets to work on them in the first place.
