Ray Kurzweil

Ray Kurzweil grew up in Queens, New York, in a family of musicians and artists. At 14 he built his first computer, and at 17 he won the U.S. national science prize with a program that composed original music. Since then, across more than five decades, he has lived convinced of an idea most people dismissed as science fiction: technology doesn't grow linearly, it grows exponentially, and that changes everything.

His most concrete contribution came in 1976, when he unveiled the Kurzweil Reading Machine: the first device capable of reading any printed typeface aloud for blind people. Stevie Wonder was one of his first users, and their collaboration led to the invention in 1984 of the Kurzweil K250, the first music synthesizer capable of recreating the sound of a grand piano and other orchestral instruments with real accuracy. Beyond that, he was a pioneer in optical character recognition (OCR), text-to-speech conversion, and large-vocabulary speech recognition. He holds more than 21 honorary doctorates and received the National Medal of Technology from President Clinton in 1999.

In 2005 he published "The Singularity Is Near," the book that made him the most influential voice in technological futurism. In it he developed the "Law of Accelerating Returns": his theory that each technology stands on the shoulders of the previous one and accelerates at a pace no human instinct can anticipate well. In 2012 he joined Google as Director of Engineering to work on natural language processing and artificial intelligence, and in 2024 he published "The Singularity Is Nearer," updating his predictions in light of the large language models he had, in some sense, anticipated decades earlier.

To me, Kurzweil's lesson is about the courage to think exponentially when everyone tells you you're exaggerating. He doesn't always get the details right, but he gets the direction right. Every time you use an AI tool that does in seconds what used to take days, there's a little of that visionary stubbornness behind it. The next time someone tells you your expectations about AI are too high, ask yourself if you might just be thinking in a straight line.