An investigation into the epistemic status of beliefs formed on the basis of large language model outputs.
Notes on AI alignment.