We don't need to prove all humans are capable of this. We can demonstrate that some humans are, therefore humans must be capable, broadly speaking
Until we see an LLM that is capable of this, then they aren't capable of it, period
Until we see an LLM that is capable of this, then they aren't capable of it, period
That's the difference. AI cannot be held responsible for hallucinations that cause harm, therefore it cannot be incentivized to avoid that behavior, therefore it cannot be trusted
Simple as that