Innovation in science and engineering is often constrained by costly experiments and simulations, complex systems, and data that is limited, biased, or unrepresentative. Physics AI addresses these challenges by combining data-driven AI with domain knowledge and first-principle models from physics, engineering, and the natural sciences.

Rather than replacing established simulation and modelling tools, Physics AI integrates them with data and machine learning to create more data-efficient, physically consistent, and generalizable models. This can accelerate computational workflows, enable faster exploration of design spaces, and improve predictions where experimental data is scarce.

Through selected techniques and recent research examples, this session will highlight how Physics AI can create new opportunities for innovation, as well as the interdisciplinary and computational challenges involved in bringing these methods into practice.

Key takeaway: Physics AI can enable faster and more predictive innovation by bringing together data, physical knowledge, and computational models, particularly where experiments are costly and data is limited.

This session is part of DTU's Industrial AI Program: From Artificial Intelligence to Industrial Intelligence at Digital Tech Summit 2026. Through a series of connected sessions, DTU brings together leading research and industry perspectives to explore how AI is reshaping products, processes, and operations across the industrial landscape.

Partners: DHI

Partners (the overall Industrial AI program): AI Denmark and AI Boost