EveryMatrix CTO Pablo Jansen: local AI models can give iGaming clients data sovereignty

EveryMatrix is taking a measured approach to artificial intelligence, according to Chief Technology Officer Pablo Jansen. Speaking to SBC Media at the company's Copenhagen headquarters, he said the aim is to understand what lies behind the industry hype, including the differences between commercial and local models.

Jansen said EveryMatrix is currently in a phase of getting to grips with local models. In his view they can help the company secure data sovereignty for its clients, because systems can run locally rather than depending on outside infrastructure. "The more local the model, the more control our partner has over the data," he said.

The distinction he draws is between models run on a client's own platform and those that operate on remote cloud servers. The past year has seen a sharp rise in the number of AI products and models offered by different providers and platforms, and Jansen said the company tries to understand the range of options and the wider landscape in full.

EveryMatrix already uses AI extensively across its business. Applications named include trading, quantitative modelling and pricing, the development of its bet builder product, and the assessment of user data and preferences.

Looking ahead, Jansen expects more secure guardrails between datasets. He said the company is working towards a model in which AI is hosted either in its own data centres or in the cloud, with the client deciding which guardrails apply. That approach would leave operators and partners with greater say over how their data is handled as AI use spreads across the iGaming sector.