The reasoning
AI produces a large range of outputs, depending on the prompt, model quality, and numerous other parameters. People using the Texas Slopshooter fallacy draw together the bad outputs (like malformed images or buggy code) and use that to draw an inference of AI as a whole; claiming this is what AI typically produces while ignoring or dismissing stronger examples.
Example
Stating, “Look at these terrible AI images I found, I can't believe anyone could ever find this usable!” when all the examples are cherry-picked from old AI image models.
