> The software does not automatically fail a student, it only flags for further human review. Seems like it’s doing its job. That is it's intended use, yes, bu

> The software does not automatically fail a student, it only flags for further human review. Seems like it’s doing its job. That is it's intended use, yes, but it does not appear to be how it's being used. The article even says those 3,000 people were sent orders to prove they did not cheat. That sure seems like you fail unless you can prove the software wrong. Unless you're claiming that they had human review of those and eliminated some, in which case the software flagged over 1 in 3 people for cheating. Either way, the software is bad. At the very least, there's a very high false positive rate. You would have to be really sure you're going to fail the bar, or absurdly stupid, to cheat at the bar. I'm sure being caught cheating is effectively a lifetime ban from being a lawyer, so you've wasted all that time in school. I would be utterly stunned if anywhere close to 1 in 3 law students cheated on the bar, and if that's the case, it should immediately trigger an ethics review by the bar.