Post

My newest gestating idea on what AI is doing to brains involves baseball. If you were trying to put together a winning baseball team — or even just trying to understand baseball better — is that possible knowing only the state of the game before every at bat and the results of those at bats?

I think there's still a strain of thought out there that "if I know 'the statistics' I can understand baseball really well". Sure, yeah, to a point.

But if you discuss *which* statistics matter for long enough, you'll eventually find, in the modern context, you're actually talking about how to measure the process of being good at baseball, and that requires knowing how baseball works on the ground, and how it's taught to players.

All that's to say that my professional experience with "AI" is largely people seeking ways to make their lives easier and almost invariably giving up on knowing how things work. To bring it back to baseball, you do NOT need to know about pitcher spin rates to enjoy the sport, BUT...

For regular office workers, eg: you don't know how to extract certain data from a spreadsheet? LEARN HOW TO DO IT. Don't just ask the machine to do it without knowing what the machine should do to get the info you need! In doing this task, you're also likely engaging with the material more deeply.