TechnologyPhoto: Quanta MagazineMoritz Hardt Calls for Scientific AI Benchmarking Reform
Moritz Hardt, author of a new book on AI benchmarks, argues that current standardized tests for artificial intelligence are fundamentally flawed. However, he believes these issues are not unfixable and advocates for transforming AI benchmarking into a rigorous scientific discipline to improve reliability and accuracy.
Quanta Magazine·




