Inverse intent has a way of disestablishing negligence. If you believe something that is false, and do harm by expressing that belief, then you are still prote

Inverse intent has a way of disestablishing negligence. If you believe something that is false, and do harm by expressing that belief, then you are still protected by your right to religious liberty. In this case, the belief is that coherent language is always constructed using symbolic logic and intent. LLMs do not categorize symbols, perform logic, or follow explicit intent; yet they are able to generate coherent language. GPT's authors have utterly failed to present this distinction. Was that failure intentional, or the result of their own belief? The result is as ambiguous as the "truthiness" of GPT's output: it can't be reasonably used as evidence for intent.