Who should be responsible when AI causes harm? How do we protect people’s data? Who owns what AI creates? How should AI systems be governed