Making Intelligence — Ethics, IQ, and ML Benchmarks

Co-presented with Leif Hancox-Li. In person on November 28 2022, and virtually on December 5 2022.

Poster presentation of our work-in-progress research paper that draws on feminist scholarship about IQ to shed lights on overlooked areas of ethical risk for ML benchmarks. The current version of the paper is available on arXiv.

Paper Abstract: In recent years, ML researchers have wrestled with defining and improving machine learning (ML) benchmarks and datasets. In parallel, some have trained a critical lens on the ethics of dataset creation and ML research. In this position paper, we highlight the entanglement of ethics with seemingly technical'' or scientific’’ decisions about the design of ML benchmarks. Our starting point is the existence of multiple overlooked structural similarities between human intelligence benchmarks and ML benchmarks. Both types of benchmarks set standards for describing, evaluating, and comparing performance on tasks relevant to intelligence – standards that many scholars of human intelligence have long recognized as value-laden. We use perspectives from feminist philosophy of science on IQ benchmarks and thick concepts in social science to argue that values need to be considered and documented when creating ML benchmarks. It is neither possible nor desirable to avoid this choice by creating value-neutral benchmarks. Finally, we outline practical recommendations for ML benchmark research ethics and ethics review.