Recent projections of energy intensity predict a more rapid decline in intensity than has occurred in the recent past. To assess how well such projections have performed in the past, I assess the accuracy of the business as usual energy intensity projections embedded in the annual World Energy Outlook (WEO) produced by the International Energy Agency since 1994. Changes in energy intensity depend on economic growth, and historical errors in projecting energy intensity can partly be explained by errors in projecting the rate of economic growth. However, recent projections of the elasticity of energy intensity with respect to economic growth probably overstate the likely future reduction in energy intensity even if economic growth is projected accurately. This could be because energy efficiency policies are not implemented as effectively as expected or because the economy-wide rebound effect is larger than modeling assumes.