AI Should Not Be An Imitation Game: Centaur Evaluations

06/04/2025

Benchmarks and evaluations are central to machine learning methodology and direct research in the field. Current evaluations commonly test systems in the absence of humans. This position paper argues that the machine learning community should increasingly use centaur evaluations, in which humans and AI jointly solve tasks.

Centaur Evaluations refocus machine learning development toward human augmentation instead of human replacement, they allow for direct evaluation of human-centered desiderata, such as interpretability and helpfulness, and they can be more challenging and realistic than existing evaluations. By shifting the focus from automation toward collaboration between humans and AI, centaur evaluations can drive progress toward more effective and human-augmenting machine learning systems.

Authors

Andreas Haupt

Digital Fellow

Andreas Haupt is an AI Institute Fellow-in-Residence at Schmidt Sciences, a Digital Fellow at the Stanford Digital Economy Lab, and a Technology and Human Rights Fellow at the Harvard Kennedy School. He studies how algorithmic systems represent and are evaluated against diverse human preferences, using methods from microeconomic theory, structural econometrics, and reinforcement learning. His work has been published in AI venues such as NeurIPS and ICML, in computer science venues such as ACM EC, ACM RecSys, and ACM FAccT, and as a lead article in the American Economic Review.

Read more

Erik Brynjolfsson

Jerry Yang and Akiko Yamazaki Professor

Erik Brynjolfsson is one of the world’s leading experts on the economics of technology and artificial intelligence. He is the Jerry Yang and Akiko Yamazaki Professor and Senior Fellow at the Stanford Institute for Human-Centered AI (HAI), and Director of the Stanford Digital Economy Lab. He also is the Ralph Landau Senior Fellow at the Stanford Institute for Economic Policy Research (SIEPR), Professor by Courtesy at the Stanford Graduate School of Business and Stanford Department of Economics, and a Research Associate at the National Bureau of Economic Research (NBER).

One of the most-cited authors on the economics of information, Brynjolfsson was among the first researchers to measure productivity contributions of IT and the complementary role of organizational capital and other intangibles.

Read more