paper / nate-soares / May 11 / failedViews on AI Existential Risk Before and After a Public Event at Harvard University