Comment by delichon
9 hours ago
How long before we start building detailed models of each judge trained on all of their legal output, and then test various legal theories against those judge models in virtual moot court? Craft each pitch to the legal idiosyncrasies of the batter. I assume that real lawyers do this routinely and could use a simulator.
I feel like this is on a different trajectory than what LLM-based tech does. This type of individualistic extrapolation is one of the things they're really bad at, in my experience.
This is already a thing to some extent. I don’t recall the products that do it, but judge profiles def exist and are taken into account .
What worries me is the step after this. If god forbid this proves successful and models accurately predict specific outcomes, people will start to ask whether the solution to AI slop lawsuits is to do the judging with AI too.