We were looking at acquiring a small AI startup. Used Elicit to do a literature review and check their claimed "novel" methods.
Found it useful for pulling up relevant papers fast. Saved a few days of manual search. The summaries are decent for a high-level scan.
But it missed key rebuttals and critical papers that our in-house expert knew about. The consensus view it presented was skewed toward older, more cited papers, which made the startup's claims look weaker than they actually were. It's a filter bubble of citations, not ground truth.
For due diligence, it's a starting point, not a finish line. You still need someone who knows the field to spot the gaps. The cost adds up quickly for deep searches, and you'll pay for those missed papers anyway when you have to go find them manually.