Detecting AI use

There’s software specifically designed for the purpose of sussing out whether AI may have been used in the creation of a piece of writing. There are many different services allegedly able to do this task, which seems to consist in taking a bit of writing and seeing if it contains constructions or tendencies which the program alleges are favorite tendencies of AI. Pangram is by the sound of it the most accurate, but I remain unsure that can we reliably distinguish writing done by an AI chatbot from writing done by a human. Seems to me it’s almost like left-wing politics or religion — you either believe in it or you don’t. I don’t know enough to have an informed position on this question: I just tend to believe testing is pointless. I mean, who thinks anything bad will happen to them if they inadvertently read a paragraph written by AI? (Or even advertently? Or even a book?) There’s plenty of bad writing by live people — and I dare say some reasonably good writing by AIs. What’s the…

Commenting on Bubbles has been removed. Why? If you want to talk about this post, write to the author, use the comments on their blog if it has them, or reply with a "RE:" post on your own blog. Suggestions? feedback@bubbles.town