Greg Reese Report - AI Governance and The Agentic State
Aired: 2025-07-31
Duration: 05:09


=== Agentic State Explained (04:44) ===

[00:00:06]  The agentic state is a psychological condition where individuals see themselves as agents executing the wishes of an authority figure.
[00:00:15]  This concept was explored in Stanley Milgram's obedience experiments, where participants administered painful electrical shocks to others under the direction of an authority figure.
[00:00:26]  Volunteers were told they were taking part in scientific research to improve memory.
[00:00:31]  Would you open those and tell me which of you is which, please?
[00:00:36]  Teacher.
[00:00:36]  Learn.
[00:00:39]  Separated by a screen, the teacher would ask the learner questions in a word game and administer an electric shock when the answer was incorrect.
[00:00:48]  He was told to increase the voltage with each wrong answer.
[00:00:52]  Cloud.
[00:00:54]  Horse, rock, house.
[00:00:57]  Answer.
[00:00:59]  Wrong.
[00:01:01]  150 volts.
[00:01:04]  Answer, horse.
[00:01:07]  Experiment one.
[00:01:08]  That's all.
[00:01:08]  Get the audio here.
[00:01:10]  These experiments showed that the agentic state leads to a loss of individual responsibility.
[00:01:16]  Those involved believe they are simply following orders and doing the right thing.
[00:01:21]  There was a time, I suspect, when men and women could give a fully human response to any situation, when we could be fully absorbed in the world as human beings.
[00:01:31]  But more often now, people don't get to see the whole situation, but only some small part of it.
[00:01:38]  There's a division of labor, and people carry out small, narrow, specialized jobs.
[00:01:44]  And we can't act without some sort of direction from on high.
[00:01:49]  I call this the agentic state.
[00:01:52]  The individual yields to authority and in doing so becomes alienated from his own actions.
[00:01:58]  The person has a choice.
[00:02:00]  He or she chooses to become a juntic.
[00:02:04]  But once you assume the role, it's almost impossible to go back.
[00:02:08]  Published in May of this year by the Global Government Technology Center in Berlin, a proxy of the World Economic Forum, the Agentic State white paper outlines how Agentic AI will revamp 10 functional layers of public administration.
[00:02:25]  The white paper states that, in an era increasingly characterized by polycrisis, interconnected and cascading shocks ranging from pandemics and extreme weather events to cyber-physical attacks, financial instability, disinformation campaigns, and even conventional warfare, traditional crisis management models are under strain.
[00:02:46]  Threat actors are already adapting.
[00:02:48]  With AI, they can automate, scale, and personalize attacks at unprecedented speed.
[00:02:54]  Governments, by contrast, are often still operating with institutional reflexes shaped for a slower, more linear world.
[00:03:03]  One of the main proposals is something called simulation infrastructure, where every individual is given a virtual twin, and everything we do will be analyzed and simulated with the idea that these simulations can be used for knowing the future actions of an individual or a group.
[00:03:20]  The report claims that when a crisis begins to unfold, AI initiates the first steps in crisis response before human in-the-loop structures have time to react.
[00:03:32]  These agents will work alongside increasingly autonomous physical systems such as drones and robots, forming the backbone of a responsive adaptive crisis infrastructure.
[00:03:44]  Co-founder of Oracle AI, Larry Ellison, explains how this might look in the near future.
[00:03:51]  The police will be on their best behavior because we're constantly recording, watching and recording everything that's going on.
[00:03:59]  Citizens will be on their best behavior because we're constantly recording and reporting everything that's going on.
[00:04:07]  And it's unimpeachable.
[00:04:10]  It's not people that are looking at those cameras.
[00:04:12]  It's AI that's looking at the camera.
[00:04:13]  No, no, no.
[00:04:14]  You can't do this.
[00:04:15]  A drone goes out there.
[00:04:16]  It goes there way faster than a police car.
[00:04:19]  There's no reason for, by the way, high-speed chases.
[00:04:22]  You shouldn't have high-speed chases between cars.
[00:04:24]  You just have a drone follow the car.
[00:04:26]  I mean, it's very, very simple.
[00:04:28]  And the new generation of autonomous drones.
[00:04:34]  And who are you, little fellow?

=== Funded by Substack (00:33) ===

[00:04:39]  Come to show them where I am.
[00:04:43]  Not nice.
[00:04:48]  Greg Reese reporting.
[00:05:04]  The Reese Report is now fully funded by my Substack subscribers.