AI for Realistic and Reproducible Science and Engineering

AI is a cornerstone technology of our era, driving notable discoveries like AlphaFold and ChatGPT. Motivated by these real-world cases, there is a growing demand for similar efforts in science and engineering. Despite the proliferation of case studies showcasing the broad impact of AI, the number of examples that demonstrate the practical benefits of AI in realistic, large-scale scientific and engineering problems—particularly those demonstrating reproducibility — remains scarce. This scarcity poses a risk: corresponding communities might hold unrealistic expectations of AI's capabilities, not fully aware of the underlying challenges. The aim of this workshop is to address this by highlighting and demonstrating the profound ramifications of AI within the domains of science and engineering, with realistic case studies drawn from large-scale experimental facilities, such as photon and neutron sources, industries, and research laboratories. By co-locating this workshop with ISC, we provide an excellent opportunity for the audience to complement their knowledge by attending general events within the theme of ISC. Our program covers various topics, emphasising realistic applications of AI that are both reproducible and scalable. It promises to be an informative platform for emerging researchers and a forum for establishing AI best practices within the scientific and engineering domains.

---

Contributors:

• Kirstine Dale