> would call out the AI hype bubble
Which is what it is by describing it as a tool needing thousands of dollars and years of time in learning fees while being described as "replaces devs" in an instant. It is a tool and when used sparingly by well trained people, works. To the extend that any large statistical text predictor would.