I’m as excited about LLMs as the next guy, but I am sort of confused about stuff like this.<p>From the article:
“One of Casetext's key products is CoCounsel, an AI legal assistant launched in 2023 and powered by GPT-4 that delivers document review, legal research memos, deposition preparation, and contract analysis in minutes, Thomson Reuters said in a statement.”<p>I get using an LLM to generate code snippets or text, where a human developer or author might be able to pick out the good parts and resolve any syntactical mistakes.<p>But for something like contract analysis or document summary, in a profession where cases have been won and lost on something as precise as comma placement, are LLMs really useful here? Their inherent inexactitude makes me wonder.