“AI ethics” is discussed everywhere. The harder question is how you translate principles into day-to-day decisions, model governance, and fair outcomes, especially in critical infrastructure.
Tim Vermeulen, Director of Digital Strategy & Architecture at Alliander (Liander), a grid operator that distributes electricity and gas to millions of customers in the Netherlands, will talk about the evolution of AI and the ethical safeguards needed to manage the Dutch power grid responsibly. He will unpack the journey of algorithmic bias, from early origins to what it means in production today, through concrete examples from Liander’s operations:
- Efficiency versus fairness: Liander currently has more than 60 AI models in production. One model fully assesses 1-to-3 phase connections using AI, reducing the need for physical home visits. How can we ensure that such decisions remain fair and explainable?
- Strategic asset management: When an AI model recommends where to lay cables, how do we prevent it from systematically prioritizing neighborhoods with higher existing EV adoption?
- The human element: Where should boundaries be drawn when it comes to monitoring and supporting employees?
To address these complex questions, Liander has established a dedicated Ethics Board. Tim will share insights into what it takes to make such a board effective in practice.
Every KINTalk is a valuable moment where academics and practitioners meet, great opportunity to connect with others and build professional relationships.The talk will conclude with a networking session — a chance to exchange ideas, share experiences, and explore potential collaborations over drinks and snacks.
This session is particularly relevant for professionals working in AI, governance, compliance, or infrastructure. Register for the KINTalk by February 9th to secure your spot.