The Future of Quality Assurance: Side and Modl.ai Pioneer AI-Driven Testing

The landscape of video game development is undergoing a seismic shift as the industry grapples with ballooning production costs and the increasing complexity of modern titles. As development teams race to deliver bug-free experiences to a demanding player base, the bottleneck of Quality Assurance (QA) has become a primary focal point for innovation.

In a significant move to address these challenges, the services provider Side—formerly known as PTW—has entered into a binding six-month memorandum of understanding (MOU) with the Danish AI specialist Modl.ai. The partnership seeks to build a sophisticated, AI-enabled quality assurance ecosystem designed to augment, rather than replace, human expertise.

The Genesis of the Partnership

The collaboration between Side and Modl.ai represents a strategic marriage of industry experience and cutting-edge machine learning. Side, a veteran of the gaming services sector, brings a deep understanding of the global QA pipeline and the logistical challenges faced by developers. Modl.ai, a newer entrant founded in 2018 with $1.7 million in seed funding, provides the technical architecture necessary to integrate AI into the testing workflow.

Martin McBride, Chief Commercial Officer at Side, describes the partnership as a pursuit of the "holy grail" of game development. "The MOU effectively brings the best of what Side has and the best of what Modl.ai has together in a unique solution," McBride notes. "No one has really found a great solution for using AI to augment QA services yet. We believe we have found it."

Evolving Beyond Traditional Integration

Historically, AI-driven testing tools faced a significant barrier to entry: friction. Previous iterations of such technology, including earlier attempts by Modl.ai, required developers to provide full access to source code, plugins, and SDKs. This approach created a high administrative burden, often leading to technical bottlenecks and security concerns.

Christoffer Holmgard, founder of Modl.ai, explains that the new system takes a fundamentally different, "black-box" approach. "We take the product as you would give it to a QA tester, as you would eventually give it to the consumer," Holmgard states. By operating on the same hardware as a standard end-user, the system simulates human behavior by "watching" the screen and executing key processes. It generates a comprehensive dataset of gameplay events, which is then analyzed by a second AI layer tasked with identifying potential anomalies, flagging them for human review.

The Workflow: From "Game Knowledge" to Bug Reporting

The integration of this system into a development cycle follows a structured, three-phase process designed to minimize error and maximize efficiency:

1. Building Game Knowledge

Before the AI begins testing, it must "learn" the title. Developers build a foundational understanding by uploading game design documents, assets, and engaging in natural language conversations with the AI. By feeding the system context, the AI becomes attuned to the specific mechanics and design intentions of the project.

2. Targeted Execution

Once the AI understands the game, developers define the testing scope. This can range from traditional test cases to custom scenarios tailored for machine learning agents. Because the AI requires specific guidance to navigate complex game states, human QA specialists act as "architects," adapting prompts to ensure the AI focuses on the most critical areas.

3. The Human-in-the-Loop Validation

The final phase is the most critical. The system generates detailed bug reports, complete with video evidence, logs, and metadata. This output is then presented to a human QA expert, who reviews the findings. This human oversight is not merely a final check—it is a feedback loop. As testers confirm or reject the AI’s findings, the system learns from its mistakes, progressively refining its accuracy over time.

Implications for the QA Workforce

Perhaps the most controversial aspect of AI in the workplace is the fear of displacement. However, leadership at both Side and Modl.ai are framing this technology as a vehicle for professional development rather than job reduction.

Moving Up the Value Chain

Martin McBride argues that the automation of repetitive "smoke tests"—the mind-numbing, mechanical tasks performed in every build cycle—will free human testers to engage in high-level strategic analysis. "We get to move our people up the value chain," says McBride. "Instead of being a tester, they can become an AI engineer."

This shift in job description could theoretically allow companies like Side to bring jobs back to regions where rising labor costs had previously rendered certain positions unsustainable. By elevating the role from manual labor to analytical engineering, the industry can create more fulfilling career paths, potentially reducing the high turnover rates currently plaguing the QA sector.

Fulfilling Careers Through Automation

Benedikte Mikkelsen, Chief Product Officer at Modl.ai, echoes this sentiment, highlighting that human testers are currently underutilized. "There are some very tedious tests that testers are performing right now," Mikkelsen explains. "You don’t really need a human doing the same flow over and over again. That person could be doing something much more valuable, like assessing the gameplay feel or doing creative testing."

By removing the "drudgery" of repetitive verification, the partnership aims to keep talent engaged. If the repetitive work is handled by the AI, the human QA specialist remains the final arbiter of quality, focusing on nuance and creative design—areas where AI still lacks the capability to make subjective judgments.

Current Limitations and the Road Ahead

Despite the optimism, the companies remain realistic about the current capabilities of the technology. Oliver Deslandes, COO at Side, is candid about the "sliding scale" of the current tech.

"The tech right now works with more simplistic titles," Deslandes admits. "Think puzzles, card games, or titles with repeatable, linear mechanics. Anything very rapid, like high-intensity fighting games, is a little further down the line."

The goal for the current six-month MOU is to partner with developers willing to embark on a "journey" of iterative testing. For studios eager to see results and save money immediately, the technology serves as a pilot program. It is not yet a plug-and-play solution for massive, 3D open-world adventures, but rather a specialized tool for studios looking to streamline production through innovation.

Conclusion: A Collaborative Future

The collaboration between Side and Modl.ai signals a maturing of AI in the gaming industry. It moves the conversation away from the hyperbolic "AI replacing humans" narrative and toward a model of "Augmented Intelligence."

By centering the human expert as an essential component of the feedback loop, Side and Modl.ai are attempting to build a sustainable bridge between legacy QA practices and an automated future. As the industry looks toward 2026 and beyond, this MOU may serve as the blueprint for how developers, service providers, and AI can coexist to solve the industry’s most pressing efficiency challenges.

For the developers who choose to participate in this early-stage journey, the reward may not just be a faster, cleaner build, but a fundamental transformation of how they approach the very concept of game quality.