Deep Learning Opacity in Scientific Discovery
Eamon Duede
Abstract
Abstract While philosophers have focused on epistemological and ethical challenges of using artificial intelligence (AI) in science, scientists have focused largely on opportunities. I argue that this disconnect between philosophical pessimism and scientific optimism is driven by failures to critically examine the practice of AI-infused science. To appreciate the epistemic justification for AI-powered breakthroughs, philosophers must analyze the role of AI as part of a wider process of discovery. I demonstrate the importance of this with two cases from the scientific literature, and show that epistemic opacity need not diminish AI’s capacity to lead scientists to significant and justifiable breakthroughs.
Topics & Concepts
EpistemologyOptimismScientific discoveryPessimismProcess (computing)Philosophy of scienceEngineering ethicsPhilosophyPsychologyCognitive scienceComputer scienceEngineeringOperating systemExplainable Artificial Intelligence (XAI)Philosophy and History of ScienceEthics and Social Impacts of AI