I've been experimenting with submitting documents as prompts to AI for analysis, but I've noticed that it often generates content that isn't actually in the documents. For instance, when I asked it to summarize an interview article, it created quotes and topics that were completely made up. Has anyone else experienced this? Is there a way to improve the prompts to avoid this issue?
1 Answer
Yeah, this is a known issue with AI right now, especially with what's called hallucinations. It's one of the main challenges we're facing.
Thanks! Are there other AIs you would recommend that might handle this better?