Putting the “intelligent” machine in its place

It’s rare for a data scientist to be so reflective and critical on the limitations (and therefore the real opportunities) of their discipline, as Katherine Bailey has managed in a short article on Techcrunch. Recent issues around machine learning biases…