Professor Kwisthout began by addressing common misconceptions about AI. A key theme of the lecture was sustainability, not only in terms of environmental impact but also in ethical and social dimensions. Professor Kwisthout stressed that AI should be evaluated against the United Nations' Sustainable Development Goals, considering factors such as energy consumption, data privacy, and fair labor practices. He highlighted concerns about the large ecological footprint of generative AI models and the importance of developing AI systems that align with societal needs rather than simply advancing technology for its own sake.
AI to help doctors make decisions
Professor Kwisthout’s own research focuses on using AI to assist human decision-making, particularly in healthcare. He discussed the development of clinical decision support systems, which use AI to help doctors make better-informed choices in patient care with a model focused on the quality of life during and after treatment.
These systems rely on so-called probabilistic graphical models, such as Bayesian networks, which allow AI to process uncertainty and explain its reasoning in a transparent way. This approach contrasts with the ‘black box’ nature of many deep learning models, making AI decisions more interpretable and trustworthy.
Sustainable and ethical AI
In his concluding remarks, Professor Kwisthout emphasized that AI should not be seen as a replacement for human judgment but as a tool to enhance it. He advocated for AI systems that are sustainable, ethical, and designed to empower users rather than exploit them. His work continues to explore how AI can be integrated responsibly into healthcare and other decision-making domains.
To learn more about Professor Kwisthout’s vision for sustainable AI, watch the full lecture here: