Side, a seasoned services provider formerly known as PTW, has entered into a six‑month memorandum of understanding (MOU) with Denmark‑based artificial‑intelligence company Modl.ai. The agreement commits the two organisations to collaborate on building an AI‑enhanced quality‑assurance (QA) platform that blends Side’s deep testing expertise with Modl.ai’s cutting‑edge machine‑learning tools. Modl.ai, a relative newcomer to the gaming sector, was founded in 2018 and secured $1.7 million in seed financing two years later.
While Side has a long history of delivering testing services across a range of industries, Modl.ai brings a fresh perspective on how AI can automate and augment the detection of bugs in video games. According to Side’s chief commercial officer Martin McBride, the partnership “effectively combines the best of Side with the best of Modl.ai into a unique solution.” He adds that the gaming industry has been searching for a “holy‑grail” AI‑driven QA experience and that the two companies believe they have finally found it. Earlier AI‑driven QA approaches, including earlier versions of Modl.ai’s own technology, required developers to hand over full game builds or even source code to the testing firm.
Those solutions relied on integration plugins and software development kits (SDKs), creating considerable friction for developers and often becoming untenable as projects grew in size. Modl.ai’s founder, Christoffer Holmgard, explains that the new system eliminates the need for any internal access. Instead, the product is supplied to the AI exactly as a QA tester or a consumer would receive it—installed on a comparable device, run, and observed. The AI watches the game play out on screen, simulates key inputs, and records a comprehensive data set of everything that occurs.
This raw data is then handed off to a second AI layer that analyses the footage, flags potential defects, and escalates the findings to a human reviewer for validation. Holmgard describes the first stage of the workflow as “building game knowledge.” Developers upload assets, design documents, screenshots, and other relevant materials, then converse with the AI in natural language to teach it the fundamentals of the title. Once the AI has a baseline understanding, developers specify the test scenarios they wish to run.
These can range from traditional test cases to novel, AI‑friendly challenges that exploit the system’s strengths. Holmgard notes that human expertise is still crucial at this point: “You need to adapt your prompts and language to guide the AI to do the right things.” During a test run—referred to by Modl.ai as a “session”—the system generates bug reports that include textual descriptions, short video clips as evidence, and detailed log files. These reports are delivered to a human QA specialist who reviews the findings and decides how to act on them within the production pipeline. One of the persistent difficulties for AI in gaming is distinguishing a genuine bug from an intended feature.
Holmgard says that feedback from human testers helps the AI learn these nuances over time. “Prompting, pre‑built game knowledge, and continuous human feedback all contribute to fine‑tuning the model,” he explains. Side envisions a future where QA personnel evolve into AI engineers, shifting the skill set from repetitive manual testing to higher‑value activities such as prompt engineering, data analysis, and model supervision.
McBride argues that this transition can raise salaries, create new career pathways, and even bring jobs back to regions where they have previously disappeared due to low‑cost labour pressures. By moving staff up the value chain, Side hopes to retain talent longer and offer more fulfilling work. According to McBride, AI excels at handling the monotonous, repetitive tests that consume a tester’s day—smoke tests, regression checks, and other routine verifications. Automating these tasks frees human testers to focus on strategic analysis, creative testing, and deeper gameplay evaluation.
Benedikte Mikkelsen, Modl.ai’s chief product officer, echoes this sentiment, noting that many current tests are “tedious” and could be performed more efficiently by a machine, allowing humans to concentrate on assessing gameplay quality and devising innovative test scenarios. Both executives stress that AI is not a replacement for human QA staff but a tool that elevates their role.
McBride points out that testing costs remain low in certain regions, making a full AI takeover impractical. Instead, the AI acts as an assistive technology that reduces human error, accelerates testing cycles, and improves overall accuracy. “There will always be a human in the loop—that’s our ethos,” he says, describing AI as a sliding scale rather than a binary replacement. When asked about the types of games the solution can currently handle, Side’s chief operating officer Oliver Deslandes admits that the technology is best suited for simpler titles—puzzle games, card games, and other experiences with repeatable mechanics.
Fast‑paced, 3D open‑world or fighting games present greater challenges and are slated for later development. Deslandes emphasizes that the early adopters will be clients willing to experiment and pilot the system, acknowledging that the journey will involve iterative improvements.
In summary, the Side‑Modl.ai partnership aims to create an AI‑driven QA platform that combines automated bug detection with human expertise, reduces repetitive workload, and opens new career avenues for testers. By delivering a product to the AI in the same form a consumer would receive it, the system sidesteps the friction of code‑level integration and focuses on observable gameplay.
Human reviewers remain essential for interpreting AI‑generated reports, providing feedback, and ensuring that nuanced design decisions are respected. As the technology matures, Side expects to expand its applicability to more complex games, while continually refining the human‑AI collaboration that lies at the heart of its philosophy.