AI and the Everything in the Whole Wide World Benchmark

The benchmarking paradigm in machine learning is incompatible with claims to performance on underspecified general tasks

Read more here: External Link