Enhancing the Scientific Foundations of Artificial Intelligence and the Scientific Impact of Quantum Computing

Speaker: Peter V. Coveney (University College London)

Abstract:

The role of artificial intelligence within science continues to grow but should not be allowed to redefine the scientific method. Here I critically assess the impact of AI on research and show with examples that AI often lacks reproducibility, transparency, objectivity and mechanistic understanding. For productive integration in scientific and many other forms of research, we need forms of AI that are fully compatible with the scientific method and conform to its high standards of integrity. With regard to quantum computing, I will summarise the state of the art and indicate the directions in which it should be evolved for it to have the biggest impact on scientific research.

This seminar will take place in room B1.25