Hallucination is all LLMs are capable of doing. Whether those hallucinations align to truth, falseness, or neither is secondary. They're trained on text but the

Hallucination is all LLMs are capable of doing. Whether those hallucinations align to truth, falseness, or neither is secondary. They're trained on text but they don't recall text, they just hallucinate "beliefs" (heavy scare quotes) that aren't rooted in any notion of logic, just empirical statistical association (assuming the training was "successful" (again heavy scare quotes)). So LLMs hallucinate because that's what their architecture demands of them for the purpose of generating outputs. It's really that simple.