Side, a seasoned services provider formerly known as PTW, has entered into a six‑month memorandum of understanding (MOU) with Denmark‑based artificial‑intelligence specialist Modl.ai. The agreement sets the stage for a joint effort to create an AI‑enhanced quality‑assurance (QA) platform that blends Side’s deep industry experience with Modl.ai’s cutting‑edge machine‑learning capabilities. Modl.ai, a relatively new player in the gaming sector, was founded in 2018 and secured $1.7 million in seed financing two years later.
Its founders have spent the past several years refining AI tools that can assist QA teams, but earlier versions required developers to hand over entire builds, source code, or integrate a custom SDK. That approach introduced considerable friction, as developers were often reluctant to expose internal assets and the integration process became cumbersome over time, according to Modl.ai’s founder Christoffer Holmgard. The new solution, Holmgard explains, eliminates the need for any direct access to a game’s inner workings.
"We treat the product exactly as you would hand it to a QA tester or eventually ship it to a consumer," he says. "The game runs on a device that mirrors the target hardware, our system watches the screen, simulates key inputs, records every observable event, and then feeds that data to a second AI layer that flags potential bugs. Those findings are then presented to a human reviewer for validation and prioritisation." The first phase of the workflow is what Modl.ai calls "building game knowledge." Developers upload assets—such as art, level designs, or design documents—and converse with the AI in plain language.
"You can chat about your game, attach screenshots, feed design docs, and the model builds a semantic understanding of your title," Holmgard notes. This natural‑language interface allows the AI to grasp the game’s mechanics, narrative flow, and visual style without needing to parse raw code. Once the AI has a foundational model of the game, developers define the test scope. This can range from traditional test cases (e.g., verifying a menu opens correctly) to more exploratory scenarios that are specifically crafted to leverage the AI’s strengths.
"Human expertise is still crucial here," Holmgard adds. "You have to phrase prompts in a way the AI understands, guiding it to execute the right actions and look for the right outcomes." During a test run, the system records textual logs, video snippets (referred to as "evidence"), and comprehensive telemetry. After the AI has analysed the session, it produces a bug report that is handed off to a seasoned QA specialist.
The specialist reviews the findings, decides which issues merit further attention, and provides feedback such as "this is a bug," "this is intended behaviour," or "this is a design choice." Over time, this human‑in‑the‑loop feedback fine‑tunes the AI’s decision‑making, allowing it to differentiate more accurately between genuine defects and deliberate features. Side envisions this collaborative model as a pathway for QA staff to evolve into AI engineers.
By equipping testers with sophisticated tooling, the company hopes to move them up the value chain, offering higher‑paying roles and new career trajectories. "Instead of a linear ladder that ends at senior tester, we can transition people into AI‑focused positions," explains Martin McBride, Side’s chief commercial officer. "That not only raises salaries but also opens the possibility of bringing jobs back to regions where low‑cost testing has been outsourced for years." McBride also stresses that AI is intended to make QA work more rewarding, not to replace humans.
Repetitive smoke tests and routine verifications consume a large portion of a tester’s day, limiting their ability to think strategically. By automating these mundane tasks, the AI frees testers to focus on higher‑order analysis, gameplay assessment, and creative testing approaches. Benedikte Mikkelsen, Modl.ai’s chief product officer, echoes this sentiment: "The most tedious, repeatable flows can be handed to the machine, allowing humans to concentrate on evaluating game feel, balance, and innovation." Despite the optimism, both companies acknowledge current limitations. Oliver Deslandes, Side’s chief operating officer, points out that the technology is presently best suited for simpler titles—puzzle games, card games, and other experiences with repeatable, low‑speed mechanics.
Fast‑paced genres such as fighting games or expansive 3D open worlds pose challenges that the team plans to tackle in later phases. "We are looking for early adopters willing to pilot the system and help us iterate," Deslandes says. "The journey is incremental, but the potential payoff in cost savings and quality improvement is significant." In summary, the Side‑Modl.ai partnership aims to create a hybrid QA ecosystem where AI handles the heavy lifting of data collection and initial bug detection, while seasoned human testers provide the nuanced judgment that only experience can deliver. This sliding‑scale approach—where AI assistance scales up or down based on the task—embodies Side’s ethos that there will always be a human in the loop.
By marrying automation with expertise, the collaboration hopes to raise the overall standard of game testing, reduce error rates, and open new professional pathways for QA personnel worldwide.