Club Brugge builds its own AI operating system with cost controls
Club Brugge is working with Ghent-based AI company ML6 on the Club Brugge OS, the club's own operating system for building artificial intelligence into its daily work, on both the sporting and business side. The platform is designed as a central system that brings together all of the club's AI applications, connecting workflows and data from different teams.
A control tower dashboard shows which AI agents are running on the Club Brugge OS and what each one costs, in total and per session, allowing the club to weigh costs against what each agent delivers. Every agent has a cost cap that stops it before it overspends, and if an agent does not perform as expected or its costs keep rising, its owner can switch it off directly from the dashboard. ML6 accelerates Club Brugge's engineering team, ensuring the knowledge stays in-house and the club fully owns what it creates.
A concrete application is supporting customer service, where supporters can ask questions to AI agents about tickets, events and merchandising. In the future, Club Brugge wants to investigate whether the technology can help with scouting and player development. For scouting and player development, the AI would not only look at the performance of players in the last match played, but also at the evolution of players over months or years, looking at the potential of the player and the playing style of the club.
The collaboration redesigns processes from the outcome the club wants to achieve backwards, because AI opens up routes that did not exist before, connecting knowledge and data that are currently spread across departments. People remain in charge throughout, with staff deciding task by task what AI takes on and how they stay involved, while AI supports and executes as the club's people set the direction and make the decisions.
If your organisation is running multiple AI agents or tools and you want visibility into what each one costs and delivers, or if you are building a platform that needs to track and cap spending per workload, we can help you design the instrumentation and controls that fit your stack.