Energy forecasting: smoke, mirrors and artificial intelligence

This pamphlet supplements the original book, "a guide to energy forecasting...part exercise in smoke and mirrors" by exploring the potential role to be played by artificial intelligence.
Forecasting energy demand internationally remains a necessary part of industry planning despite inevitable errors. Apparent global accuracy disguises counteracting errors in constituents. The superficial objectivity from the generation of myriad numbers disguises the scope for significant subjectivity in underlying assumptions, which drive outcomes supporting pre-existing conclusions. The emergence of AI means that less time and resources will be spent on gathering and assembling forecasts. This will facilitate scenario and probabilistic planning. However AI will not remove bias in forecast outcomes.