The effect of subjectivity

[Ron]

I recently watched a video, https://youtu.be/dXQPL9GooyI from an AI/machine learning guy, which presented huge, and fascinating,  differences between searches for "novelty" vs. searches for identified objectives.  One statement from the video:  expertise is the ability to determine what is interesting from an array of results.  In this study, "novelty" was subjective.

With that in mind, will GCP in the foreseable future reach the limits of objective science?  Will there be a need for new methods of science to investigate phenomena that are difficult to replicate and perhaps are affected by subjective factors?  A hundred years ago, William James wanted to "enlarge the scope of science to include the study of phenomena that are random, non-repeatable, and dependent on universal personal capacities and dispositions."

How far are we from being at an impasse without such an enlargement of scope?  Or is that what GCP is all about, i.e. enlarging that scope, though so far using conventional statistical methods?