Hacker News new | ask | show | jobs
by ashmil 657 days ago
what if it teaches some wrong content due to hallucination ?
1 comments
sayonidroy 656 days ago
We have provided LLMs with reference materials we created ourselves. Additionally we are running validators to check the output. I agree that the LLM will still not be 100% accurate but it should work pretty well
link
lionkor 656 days ago
If you have validators check that the output is near 100% accurate, you could just cut out the AI and just use the validators as tutors.
link