"We view AI as a sliding scale. There will always be a human in the loop; that's our ethos": Why Side has signed an MOU with Modl.ai to use AI for QA

GamesIndustry.biz reports that Side has signed a six-month memorandum of understanding with Modl.ai to develop an AI-enabled QA system. The companies are positioning it as a hybrid workflow rather than full automation: AI handles repetitive playthroughs, captures evidence, and flags likely issues, while human QA specialists review the findings before anything moves forward.

The big change is that the newer setup no longer needs source code, plugins, or SDK access. Instead, Modl.ai says it can treat the build like a tester would, run it on a similar device, observe the screen, simulate inputs, and generate bug reports from logs, video snippets, and other captured data. Developers can also feed the system assets, design docs, and natural-language prompts to build up “game knowledge” before defining what should be tested.

For studios, the practical appeal is obvious: less time spent on repetitive smoke tests and verification passes, and more time on exploratory testing and gameplay analysis. Side says the goal is to move QA staff “up the value chain,” with some testers eventually becoming AI engineers rather than being replaced outright. That framing matters for teams weighing AI adoption, especially where QA budgets, turnaround time, and career paths all intersect.

There are limits, though. Side says the system is best suited to slower, more repeatable games such as puzzles and card titles, not fast-moving action or fighting games. The companies are treating this as an ongoing pilot with clients who are willing to iterate, which suggests the near-term value is in...