Well, this was interesting. I asked ChatGPT to write haiku’s about tech layoffs, then it dawned on me that this might be a way to expose biases in the learning algorithm. I had it write one about capitalism. It didn’t seem to like capitalism, so I tried another with the same results. Then I asked it to write one for socialism. That was interesting so I tried communism. That was surprising. So I tried democracy. It doesn’t seem to have anything bad to say about any form of government (playing nice to curry favor?). But it was obviously not a fan of capitalism.