[PLENARY] Opportunities for AI in the Academic Medical Environment

Artificial intelligence is advancing at a breathtaking pace. Consumer AI models trained on large and varied datasets have shown remarkable breadth of applicability. This success has translated into pathology, where recent examples include models have been trained with hundreds of thousands of whole slide images with diverse pathologies, or image-text pairs harvested from internet databases and social media. These models present new opportunities to accelerate AI development in areas where data is lacking and are an exciting direction in a field defined by sub specialization of human experts. This talk will explore these developments and other pertinent topics, including the relevance of AI model interpretability and the explainability of predictions, and the challenges of studying AI in real-world settings.