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 effort to build an AI‑enhanced quality‑assurance (QA) platform that blends Side’s deep testing experience with Modl.ai’s cutting‑edge machine‑learning technology.
Modl.ai, a relatively new player in the gaming sector, was founded in 2018 and secured $1.7 million in seed capital two years later. Its founders have been experimenting with AI‑assisted QA for several years, but earlier versions of the technology required developers to ship full game builds—or even source code—to the QA vendor. That approach proved cumbersome, as it forced studios to expose internal assets and created a heavy integration burden.
According to Modl.ai’s founder Christoffer Holmgard, the friction caused by those “integration‑based” solutions made them impractical for long‑term use. The new system that Side and Modl.ai are developing eliminates the need for any direct access to a game’s internals.
Holmgard explains that the workflow mirrors the way a human tester would receive a finished product: the game is installed on a device that is identical—or at least very similar—to the end‑user’s hardware. The AI then runs the game, watches the screen, simulates player inputs, and records every observable event. A second AI layer analyses that data, flags potential defects, and escalates the findings to a human QA specialist for final validation. The first phase of the process is what Modl.ai calls “building game knowledge.” Developers upload assets such as screenshots, design documents, and other reference material, and then converse with the AI in plain language.
This natural‑language interaction helps the system form a contextual understanding of the game world, mechanics, and intended player experience. Holmgard notes that the dialogue can include text, images, and even video snippets, allowing the AI to absorb a rich, multimodal picture of the title. Once the AI has a baseline comprehension, developers define the testing objectives.
These can range from conventional test cases—like verifying that a menu opens correctly—to more exploratory scenarios that are specifically crafted to leverage the AI’s strengths. Holmgard emphasizes that human expertise is still essential at this stage: testers must phrase prompts in a way the AI can interpret, and they must decide which aspects of the game are most appropriate for automated scrutiny.
When a test run (or “session,” as Modl.ai refers to it) completes, the system generates a comprehensive bug report. The report includes textual descriptions, short video clips that serve as evidence, and detailed log files. This package is handed over to a senior QA analyst, who reviews the findings, determines their severity, and decides on the appropriate remediation steps.
Holmgard points out that distinguishing a genuine bug from an intentional feature can be challenging for an algorithm, and that human feedback is crucial for teaching the AI the subtle nuances of game design. Over time, the AI refines its judgment through a feedback loop.
Testers label each report as a bug, a non‑bug, or an expected behavior, and the system adjusts its internal models accordingly. This iterative learning process gradually reduces false positives and improves the relevance of future detections.
Side sees the partnership as a pathway to elevate its workforce. Chief Commercial Officer Martin McBride envisions QA staff transitioning into “AI engineers,” a role that combines testing knowledge with technical proficiency in machine‑learning tools.
By moving employees up the value chain, Side hopes to offer higher salaries, create new career trajectories, and even repatriate jobs that have migrated to lower‑cost regions in the past. McBride explains that AI will take over the most repetitive, low‑skill tasks—such as running smoke tests on every new build—freeing human testers to focus on strategic analysis, gameplay evaluation, and creative test design. This shift, he argues, leads to more fulfilling careers and better retention of talent. Benedikte Mikkelsen, Modl.ai’s chief product officer, echoes this sentiment, noting that automating tedious test flows allows humans to concentrate on higher‑order activities like assessing player experience and devising innovative testing methodologies.
Both leaders stress that AI is not a replacement for human QA but a complementary tool. McBride describes AI as a “sliding scale” where the human‑in‑the‑loop principle remains non‑negotiable. He argues that while the cost of manual testing in certain low‑wage markets is currently very low, providing testers with AI‑augmented toolsets extends their career longevity and opens up opportunities for upskilling. Regarding the types of games that can benefit from the solution today, 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, graphically intensive 3D open‑world or fighting games remain a longer‑term goal, as they demand more sophisticated perception and decision‑making capabilities. Deslandes emphasizes that early adopters will likely be studios willing to pilot the technology and collaborate on its evolution, even if their immediate expectations focus on cost savings and faster turnaround. In summary, the Side‑Modl.ai collaboration aims to create a hybrid QA workflow where AI handles bulk, repetitive testing while seasoned human analysts provide oversight, contextual judgment, and continuous training for the algorithms.
The six‑month MOU will serve as a proving ground for this approach, with the potential to reshape how the gaming industry conducts quality assurance, enhance employee career paths, and ultimately deliver more reliable, polished games to players worldwide.