AI is great. Midwits with AI are dangerous. I've been saying for a long time that the failure mode for AI isn't the AI itself, but the humans using it, and the

AI is great. Midwits with AI are dangerous. I've been saying for a long time that the failure mode for AI isn't the AI itself, but the humans using it, and the better the AI gets, the more I think that's borne out.