Electric vehicles registered in the US

Here we have two things that have absolutely nothing to do with one another—the number of people deciding their commute should be powered by the slow release of electrons, and the number of researchers deciding that teaching computers to think is a publishable endeavor—moving in almost perfect synchronisation for thirteen years. One wonders if there is some cosmic clipboard somewhere ticking boxes, or whether humanity simply decided that 2011 was the year to start believing in two different futures simultaneously. The correlation is 0.982, which is the sort of number that makes statisticians weep into their morning coffee.

The truth, which is slightly less poetic, is that both electric vehicles and machine learning papers are riding the same three waves: cheap computing power getting cheaper, venture capital looking for anywhere to land its considerable weight, and a growing sense that climate change and artificial intelligence are probably important. Between 2011 and 2023, global GDP per capita climbed, battery costs fell like a stone (from roughly $1,100 per kilowatt-hour to $140), and suddenly both the EV market and academic AI research had oxygen. It's a bit like watching two plants grow in the same sunlit room—they're not growing because of each other, but because someone finally opened the curtains.

What we are looking at, then, is not causation but rather two different branches of the same technological optimism, both of them sprouting from the same soil of computational abundance and cultural momentum. This should make us slightly suspicious of our pattern-seeking brains, which are now convinced that electric vehicles cause papers about neural networks and vice versa. The real variable is probably something far less visible: our collective decision to believe in technology as the answer to two separate problems.