Agents act on their own. If the hammer looked at what you wanted nailed and said, "Sorry, Dave, I can't do that."

There are degrees of autonomy, of course, and not all noncompliance is bad. Same as with humans; biological agents.

So, the difference is that you need to delete a few bad training runs?

But it seems much easier to realign an agent until it complies. Or ditch it and grab a new one.