International Conference on AI in Work, Innovation, Productivity and Skills

Can AI Read and Count? Assessing Computers with an OECD Test

Feb 23, 2022 | 4:00 PM - 5:00 PM


Understanding the impacts of AI and robotics fundamentally rests on an understanding of the technology’s capabilities. In 2016, an OECD/CERI pilot project assessed computers’ capabilities using an international test of literacy and numeracy competences. This test assesses skills in the adult population and is linked to information about work and other adult activities. The OECD has just repeated this evaluation to gauge AI development over the past five years. This session will present the initial findings of the evaluation to compare how the results changed from 2016 to 2021. The participants will then discuss some of the questions on the test and the ways that recent developments in AI would affect its performance on these questions. Moderator: Art Graesser, Professor of Psychology and Intelligent Systems, University of Memphis Panellists: • Stuart Elliot, Senior Analyst, Directorate for Education and Skills, OECD • Ken Forbus, Walter P. Murphy Professor of Computer Science and Professor of Education, Northwestern University • Yvette Graham, Assistant Professor, Trinity College Dublin • Chandra Bhagavatula, Senior Research Scientist, Allen Institute for AI (AI2) The live session time above reflects your computer's local time zone. The session will be recorded and available on replay the day after the live stream.

Presented by

More sessions of interest