That's not the problem, that's an alternative and possibly complementary solution. The problem, as we both broadly define it, is hiring people that will help the business achieve its goals. Yes, I did not apply this method over the entire employee life cycle in this one post.
The method of frequent firing as feedback is far from optimal and not even feasible in smaller organizations. That is the context in which I'm operating and thinking about this problem.
With respect to the particular element of confirmation bias referenced in the first sentence, and repeating the second, hiring based on interest is a method that will seek to truthfully answer the question, "is this person going to solve the problem we want to solve?" rather than effectively allowing an employer to avoid a costly mistake.
Your comments about product development are not addressing anything in the post and could be delivered to a wall with similar effect.