The agentic AI era promised to replace humans with autonomous systems. The evidence says that project failed on two fronts: the technology has not delivered reliable replacement, and the public is rejecting the premise even where it partially works.

This episode is an audio overview of the white paper "Why Agentic AI Was Always Going to Fail," published at basilpuglisi.com. The paper introduces the Named-Human Test, a single sorting question that classifies any AI system: is there a named human with binding authority over the output who answers through moral, professional, civil, or criminal accountability?

The episode covers:

The production evidence: agents completing roughly 24 percent of workplace tasks, multi-step accuracy collapsing by nearly half ...