I think you are on to something. But I don't think the day X will ever come. It is more like the divide between the best software and the average is going to gr

I think you are on to something. But I don't think the day X will ever come. It is more like the divide between the best software and the average is going to grow and grow. We will see the amplitude of (critical) failures increase further and their occurrence frequency rising as well. It is and will continue to be evaluated statistically: The cost of actually fixing them vs. the cost of just letting them run wild.