← Back to context Comment by retrochameleon 2 months ago Hopefully in decades hallucinations will be largely solved. 7 comments retrochameleon Reply tempfile 2 months ago Hallucinations are an essential feature of the technology, they cannot be "solved". You may as well hope that we solve the halting problem. timacles 2 months ago its just as likely hallucinations will only get worse because their source data will be riddled with hallucinations eru 2 months ago You can't hallucinate a working lean proof. darkwater 2 months ago But you can hallucinate everything else. tempfile 2 months ago You absolutely can. How do you know your "working lean proof" actually proves the theorem you intended it to? 2 replies →
tempfile 2 months ago Hallucinations are an essential feature of the technology, they cannot be "solved". You may as well hope that we solve the halting problem.
timacles 2 months ago its just as likely hallucinations will only get worse because their source data will be riddled with hallucinations eru 2 months ago You can't hallucinate a working lean proof. darkwater 2 months ago But you can hallucinate everything else. tempfile 2 months ago You absolutely can. How do you know your "working lean proof" actually proves the theorem you intended it to? 2 replies →
eru 2 months ago You can't hallucinate a working lean proof. darkwater 2 months ago But you can hallucinate everything else. tempfile 2 months ago You absolutely can. How do you know your "working lean proof" actually proves the theorem you intended it to? 2 replies →
tempfile 2 months ago You absolutely can. How do you know your "working lean proof" actually proves the theorem you intended it to? 2 replies →
Hallucinations are an essential feature of the technology, they cannot be "solved". You may as well hope that we solve the halting problem.
its just as likely hallucinations will only get worse because their source data will be riddled with hallucinations
You can't hallucinate a working lean proof.
But you can hallucinate everything else.
You absolutely can. How do you know your "working lean proof" actually proves the theorem you intended it to?
2 replies →