Synopsis
“Correlation is not causation.” This mantra, chanted by scientists for more than a century, once led to a virtual prohibition on causal talk. Today, with causal analysis taking center stage in AI and other applications, the old taboos are a thing of the past.
The causal revolution, instigated by Judea Pearl and others, established causality on a firm scientific basis. Pearl showed how intuitive models can interface with data to help answer causal questions, such as whether a drug will cure an illness or, counterfactually, whether an illness would have persisted absent the drug. These models provide a blueprint for causal AI to learn and take responsibility for its mistakes.
Updated with a new preface and summary of recent advances, The Book of Why outlines the basics of cause and effect for lay readers and scientists alike, providing tools to explore the observable world and worlds that might have been.