TellUnit testing with Java, day by day · Day 11
If an AI writes tests from an existing Java method rather than from the refund requirement, which fragile evidence is it likely to preserve?
If an AI writes tests from an existing Java method rather than from the refund requirement, which fragile evidence is it likely to preserve?
Share this tell
The image has the link printed on it, so it still leads back here.
This is one tell — the giveaway that AI is bluffing. It comes from Unit testing with Java, day by day, a short daily course on Noesa. Anyone can use AI; few can tell when it's wrong.