PERSON
Michael I. Jordan
The most influential computer scientist of his generation, who built the mathematical foundations of modern machine learning—probabilistic graphical models, variational inference, Latent Dirichlet Allocation—and then spent his public career insisting that the revolution everyone was celebrating had not actually happened yet.
There is a particular authority that belongs to the person who built the thing everyone else is mythologizing, and Michael I. Jordan possesses it at the highest possible level. By citation analysis he has been identified as the most influential computer scientist in the world; his doctoral students include many of the leading figures of contemporary AI; and the mathematical machinery he helped develop—
probabilistic graphical models, variational inference, the topic model that taught a generation to find hidden structure in text—is so deeply embedded in the field that it has become invisible, the water the field swims in. And yet, when the world began speaking of an intelligence awakening in silicon, Jordan’s response was patient, almost stubborn correction: we are not doing what we think we are doing, and pretending otherwise costs us dearly. His central diagnostic is that the term “artificial intelligence” has become a source of confusion not only for