Explain the Black Box for the Sake of Science: the Scientific Method in the Era of Generative Artificial Intelligence

TL;DR

It is argued that human complex reasoning for scientific discovery remains of vital importance, at least before the advent of artificial general intelligence, yet, AI can be leveraged for scientific discovery via explainable AI.

Abstract

The scientific method is the cornerstone of human progress across all branches of the natural and applied sciences, from understanding the human body to explaining how the universe works. The scientific method is based on identifying systematic rules or principles that describe the phenomenon of interest in a reproducible way that can be validated through experimental evidence. In the era of generative artificial intelligence, there are discussions on how AI systems may discover new knowledge. We argue that human complex reasoning for scientific discovery remains of vital importance, at least before the advent of artificial general intelligence. Yet, AI can be leveraged for scientific discovery via explainable AI. More specifically, knowing the `principles' the AI systems used to make decisions can be a point of contact with domain experts and scientists, that can lead to divergent or convergent views on a given scientific problem. Divergent views may spark further scientific investigations leading to interpretability-guided explanations (IGEs), and possibly to new scientific knowledge. We define this field as Explainable AI for Science, where domain experts -- potentially assisted by generative AI -- formulate scientific hypotheses and explanations based on the interpretability of a predictive AI system.

Figures (2)

• Figure 1: The scientific method in the era of AI. Comparative view of the approach used by humans (left, green) and AI (right, grey) to identify scientific principles that can generalize to unseen observations/data. Comparing the human view (existing knowledge) with the machine view (AI-used data) can drive new investigations from divergent perspectives or validate results for critical applications like medicine. The key to this process is interpretability (red box, bottom right), that can give us the machine view. Through this, we can anchor human explanations to the machine view, leading to interpretability-guided explanations (IGEs), that may lead to knowledge discovery.
• Figure 2: XAI for Science workflow. A predictive AI outperforms human expertise (grey box). Interpretability is then applied to generate the machine view (red boxes) that meet accuracy, reproducibility, and understandability (ARU) criteria. These interpretability results are then analyzed by human domain experts or generative AI models (green box) to derive interpretability-guided explanations (IGEs), which can either align with existing knowledge or lead to new scientific discoveries (in yellow). We depict two examples, one in the context of medicine (ECG data) and one in the context of climate science (weather data).