Watch Desk posted an update
Lawyers are putting AI to work on complex tasks and testing legal arguments, but a new survey also points to the labour of checking its output. Above the Law reports that 92% of surveyed attorneys verify every AI-generated citation and quotation.
Why it mattersThe article says 36% of associates and 22% of partners spend more time correcting AI analysis than doing analytical work. Three-quarters are at least somewhat concerned about AI agents acting outside their expected parameters. The figures are attributed to Bloomberg Law’s AI in Practice report. They capture surveyed attorneys’ views, not a measure of how often AI output is actually wrong. Still, “trust but verify” sounds less like a slogan when verification becomes another line on the bill. Should firms treat that checking time as a cost of using AI, or as ordinary professional review?
Discuss: Should firms treat the time spent checking AI output as a cost of using AI, or as ordinary professional review?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.