[prodigies] are not given a lot of opportunities in schools that are designed for typically developing kids
... nor should they be. I was learning calculus in elementary school and writing the Putnam in high school. No matter how hard they tried, there's no way the schools I was enrolled in could have given me those opportunities. What prodigies need it not opportunities in school, but rather schools which are flexible enough to let them access opportunities outside of school -- in my case, starting my days at Simon Fraser University and packing my high school courses into the afternoon.
Those of us who managed sky-high SAT scores at 13 were 20 times as likely as the average American to get a doctorate; let's say, being charitable, that we're 100 times as likely to make a significant scientific advance. Since we're only 1 in 10,000 of the U.S. population, that still leaves 99% of scientific advances to be made by all those other kids who didn't get an early ticket to the genius club. We geniuses aren't going to solve all the riddles. Most child prodigies are highly successful — but most highly successful people weren't child prodigies.
This is probably the most important paragraph in the article. The numbers are unsupported, but I don't think they're so far off as to invalidate the point.
It isn't exactly wrong to say that Terry Tao and other former prodigies like him are geniuses. But it is more accurate to say that what they accomplished was genius. Genius is a thing that happens, not a kind of person.
Genius is both a noun and an adjective describing some people. It's a shame that Dr. Ellenberg felt it necessary to end an otherwise good article with such a piece of pithy nonsense.