against the grain opinion: once you reach a certain level, it’s actually a lot slower to get gpt to generate code for you and then go through the slog of proofreading its obvious errors, and scary non-obvious ones. i gave it a try for a period of time because of the praise it was receiving for automation, but i wouldn’t really trust it for non-basic code gen. great for explainers though.

not scared. it’s super unlikely all of your backups will fail at once assuming they’re not all the same method attached to the same hardware

AI, ironically. specifically, my data being used to train big models without my consent (or, sneakily un-opt-out-able consent by using services online). it’s been accelerating the enshittification of everything by a ton. self-hosting has been a nice reprieve honestly.