Using AI for Research

I briefly discussed this on Twitter with Jack Clark and Sam Arbesman, and the idea has still been bouncing around my head —  are there types of problems or fields more generally that we would be better off designing AI to solve (or explore the space) rather than trying to solve everything “by hand?”

Read More »


Xerox PARC Was Never a Failure

In the 1970s, Xerox PARC was one of the most dominant innovation entities in the world.  Its track record was astonishing: PARC lab members developed the first personal computer with a graphical user interface, one of the earliest versions of the mouse, laser printing, and Ethernet LAN, among many other inventions.  Yet, PARC is widely regarded as one of the greatest failures in Silicon Valley history, because Xerox failed to bring any of these products to the mass market (despite the obvious, and historically vindicated, commercial potential).

Read More »