Side, a veteran services provider formerly known as PTW, has entered into a six‑month memorandum of understanding with Denmark‑based artificial‑intelligence specialist Modl.ai. The agreement sets the stage for a joint venture that will integrate Modl.ai’s AI capabilities into Side’s quality‑assurance (QA) processes, creating a hybrid system that blends machine efficiency with human expertise.
### Background and Strategic Fit Modl.ai, founded in 2018, broke into the gaming sector after securing $1.7 million in seed funding in 2020. Though relatively new, the company has been experimenting with AI‑driven testing tools for several years.
Side, on the other hand, has a long history of delivering outsourced services to game developers and publishers. According to Side’s chief commercial officer Martin McBride, the MOU “effectively brings the best of what Side has and the best of what Modl.ai has together in a unique solution.” He adds that the partnership aims to address a long‑standing industry desire for a reliable, AI‑enhanced QA workflow—a “holy grail” that many have attempted but few have mastered. ### From Integration‑Heavy to Plug‑and‑Play Earlier versions of AI‑enabled QA, including Modl.ai’s own prototypes, required developers to ship entire game builds to the testing firm.
Those builds often had to be accompanied by source code, plugins, and SDKs, creating friction and security concerns. As Modl.ai founder Christoffer Holmgard explains, that model “became difficult over time.” The new approach eliminates the need for deep code access. Instead, developers provide the same binary that a human tester would receive, install it on a device that mirrors the consumer’s hardware, and let the AI interact with the game purely through visual and input channels.
The system observes the screen, simulates key presses, records every frame and event, and then passes the resulting dataset to a secondary AI model that flags anomalies, classifies them as potential bugs, and escalates them for human review. This two‑stage pipeline preserves the confidentiality of proprietary code while still allowing the AI to learn from actual gameplay.
### Building "Game Knowledge" Before any testing can begin, developers upload assets—artwork, design documents, level layouts, and other relevant materials—to Modl.ai’s platform. The AI ingests these inputs and engages in a natural‑language dialogue with the development team, asking clarifying questions and forming a semantic model of the game’s mechanics, narrative, and intended player experience. Holmgard emphasizes that the interaction is entirely conversational: “All of this is in natural language. We just chat with the system about our game, but we’re also adding in images, or we might upload our game design documents and so on.” ### Human‑In‑The‑Loop Prompt Engineering Once the AI has a baseline understanding, developers specify the testing objectives.
These can range from traditional test cases—such as verifying UI elements or collision detection—to novel scenarios that are deliberately crafted to exploit the AI’s pattern‑recognition strengths. Holmgard notes that translating these goals into effective prompts requires human expertise: “You need some human expertise here to adapt the prompting and the language to the AI agent to make it do the right things.” This step ensures that the AI’s actions align with the creative intent of the game and that edge cases are not overlooked.
### Generating Evidence‑Based Bug Reports During a test run, the AI captures textual logs, short video clips (referred to as “evidence”), and comprehensive telemetry. These artifacts are compiled into a structured bug report and handed off to a human QA specialist for validation. The specialist decides whether the issue is a genuine defect, an intended feature, or a false positive.
Over time, the feedback loop—where humans label reports as “bug,” “not a bug,” or “acceptable design choice”—helps the AI refine its decision‑making algorithms, gradually reducing the rate of misclassifications. ### Upskilling QA Teams into AI Engineers Side envisions a future where traditional testers evolve into AI engineers.
McBride explains that this shift “moves our people up the value chain,” allowing them to command higher salaries and pursue more strategic responsibilities. By automating repetitive smoke tests and regression checks, the AI frees QA staff to focus on higher‑order analysis, gameplay evaluation, and creative testing methodologies. This transition could also reverse the offshoring trend that has seen many low‑cost testing roles migrate away from developed markets; as the work becomes more specialized, it can be relocated back to regions where skilled AI talent is available.
### Enhancing Career Fulfilment According to McBride, the reduction of monotonous tasks translates into a more satisfying career trajectory for Side’s employees. “When you knock out the repetitive tests, testers can think strategically, analyze results, and grow their expertise,” he says.
This sentiment is echoed by Modl.ai’s chief product officer Benedikte Mikkelsen, who points out that eliminating tedious, repetitive flows enables testers to concentrate on assessing gameplay quality and conducting inventive test scenarios—activities that add genuine value to the development process. ### AI as a Complement, Not a Replacement Both companies stress that AI will never fully replace human QA professionals. McBride describes AI as a “sliding scale” where the human element remains indispensable: “There will always be a human in the loop—that’s sort of our ethos.” He adds that price pressures in certain regions make it impractical to eliminate human testers entirely; instead, AI serves as a force multiplier, extending the capabilities of existing teams.
### Current Scope and Future Roadmap Side’s chief operating officer Oliver Deslandes admits that the present version of the technology is best suited for simpler titles—puzzle games, card games, and other experiences with repeatable mechanics. Fast‑paced, graphically intensive 3D open‑world or fighting games remain a longer‑term goal.
The focus, for now, is on partners willing to pilot the solution and provide iterative feedback. Deslandes emphasizes that the journey is collaborative: “If you are at the table now with us and maybe doing some pilots, then you will see where this tech goes.” ### Conclusion The Side‑Modl.ai partnership represents a pragmatic step toward integrating AI into the QA pipeline while preserving the essential role of human judgment.
By offering a plug‑and‑play testing environment, building conversational game knowledge, and creating a feedback‑rich loop between AI and experts, the collaboration aims to reduce repetitive workload, improve bug detection accuracy, and open new career pathways for QA professionals. As the technology matures, it may expand beyond modest titles to tackle more complex games, but for now the emphasis remains on delivering reliable, scalable AI‑assisted testing for developers seeking cost‑effective quality assurance solutions.