[gamerDude]

Well, they had the data around if the answer would be harmful to the students learning. AI was scored at 3.5% harmful answers and 12% of law professor answers were considered harmful.