Decision-Making Skills in an AI World: Lessons from Online Chess

Abstract

The recent surge in AI technologies places a premium on decision-making skills as a complement to the technology. Yet, little is known about how decision-making skills are developed. We theorize and provide empirical evidence of the benefits of experience – intensity, diversity and difficulty – for developing quality decision-making skills, evaluated as such relative to the performance of a mature AI – a technology with performance superior to that of humans – as a benchmark. AI advancements not only increase the importance of decision-making skills as complements, but also provide a new benchmark against which organizations will likely evaluate talent, given the alternative of seeking automation. We test our hypotheses in the context of online chess, a setting where decision-making skills are critical for performance and where a mature benchmark AI exists.