Health Ranger - Mike Adams - Data Centers and Many Worlds: Simulation Theory (Part 1)
Aired: 2026-05-07
Duration: 38:41


=== The Hidden Data Center Plan (08:36) ===

[00:00:03]  So, data centers are really striking a nerve with lots of people, with lots of concerns.
[00:00:09]  And I'm one of those people, by the way.
[00:00:11]  And I sent out a tweet a few hours ago.
[00:00:15]  I said, oh, and for whatever reason, this tweet, I posted it on X and elsewhere.
[00:00:20]  It went super viral on X and it got 400,000 views and counting, which was really odd.
[00:00:29]  But let me just read it to you, and you can see.
[00:00:30]  I said, quote, something's totally off about the number of data centers being built, over 3,000 right now.
[00:00:37]  and the sheer size and compute power they represent.
[00:00:41]  They are massively overbuilding capacity that can't possibly be met by customer demand for compute.
[00:00:48]  And customer revenues can't possibly recover the financial investment needed on these projects.
[00:00:53]  There's clearly some other plan afoot, and I don't yet know what it is.
[00:00:58]  It involves massive compute, but not merely to serve inference or hosting databases and corporate data.
[00:01:04]  There's a much larger plan at work here.
[00:01:08]  And lots of people retweeted this or replied to it.
[00:01:11]  Jimmy Doar is one of them.
[00:01:12]  Another one is author and former software engineer Rizwan Verk, who talks about simulation theory.
[00:01:22]  He said, quote, same thing happened during the dot-com days.
[00:01:25]  Check out what went wrong at Exodus, the number one web hosting company in the world at the end of the 20th century.
[00:01:32]  And yes, indeed, I remember those days very well.
[00:01:36]  I was warning people about the dot-com crash. beginning in 1998.
[00:01:40]  So I was a little bit early on that one, but you know what happened after that.
[00:01:44]  So anyway, I came to a realization that I want to share with you here about what I think the data centers are actually up to.
[00:01:56]  And it's not good, and it's going to be hard to believe.
[00:02:02]  So I'm going to present this as a theory.
[00:02:07]  You can believe the theory, you can discard the theory, you know, you can mock the theory.
[00:02:11]  You can come up with your own theory.
[00:02:13]  But this is my theory, and I'm going to back it up.
[00:02:18]  And I'm not 100% certain this is what's happening, but it seems like the most likely direction.
[00:02:24]  So let me start this.
[00:02:27]  Oh, and by the way, there's no short way to cover this topic, unfortunately.
[00:02:31]  So I'm going to do this presentation in two parts.
[00:02:34]  And you're listening to part one.
[00:02:35]  So let's start with the graphic that I put together.
[00:02:39]  It's called The Global Data Center Buildout, the planned and under construction sites worldwide.
[00:02:46]  2026.
[00:02:47]  Now, I had AI agents conduct the research on this just yesterday, and I had them build out a complete list globally of every data center that's either currently under construction or has been announced and is planned for construction anywhere in the world.
[00:03:05]  And specifically, I was looking for data center location, size in terms of power consumption, water consumption, land use in square kilometers typically, and then investment required, etc.
[00:03:19]  And this chart is the result of that because then I had another image engine map it out on a global map to show you where all these data centers are located and how large they are relative to the other data centers in terms of power consumption.
[00:03:36]  So check out this map.
[00:03:39]  So first of all, let's zoom into the United States and the number one state in the continental United States for data centers is Virginia.
[00:03:50]  Actually, Virginia.
[00:03:50]  Well, and there's a lot in Ohio, as you can see from this map, but I'm talking about existing data centers.
[00:03:56]  It's Virginia.
[00:03:57]  So let me back up and tell you there are 11,000 existing data centers in the world right now.
[00:04:06]  The 3,000 mentioned on this map are just the ones that are newly constructed or being constructed or have been planned.
[00:04:14]  And as you can see, in Piketon, Ohio, there's a SoftBank data center.
[00:04:19]  There's a lot of them in Ohio.
[00:04:22]  There's some in Texas.
[00:04:24]  And by the way, the lines on this image aren't.
[00:04:27]  Always pointing to the right place because it's an AI generated image, but the underlying data are in fact quite accurate.
[00:04:36]  So, anyway, you'll see there's lines pointing to Arizona that appear to be pointing to Seattle, etc.
[00:04:44]  But the overall point of this is that the cities that are listed are accurate and the nations that are listed are also accurate.
[00:04:53]  So, there's large data centers in Abu Dhabi, in the United Arab Emirates, there's a very large data center under construction in Shanghai, China.
[00:05:02]  There's a large one in Tokyo, Japan.
[00:05:04]  There's even one in Malaysia, etc.
[00:05:06]  You can get an overall impression, but again, pardon the geography mistakes that are part of this.
[00:05:13]  So the bottom line is that we're looking at about 190 gigawatts of power draw once these are all done.
[00:05:24]  So that's just for these new data centers.
[00:05:30]  The projected annual power draw will be over 1200 terawatt hours per year, which is a lot.
[00:05:39]  That's like 10% of all the power produced by the nation of China, just to give you a comparison there.
[00:05:46]  These data centers will take up over 1,000 square kilometers, quite a bit more, and they'll use about 15 plus billion liters of water per year.
[00:05:56]  Some of that is being siphoned from what would otherwise be neighborhoods or human households, etc.
[00:06:04]  So this is a problem.
[00:06:05]  You know, I mean, this is a growing problem, but even from a financial perspective, the question is, where's the revenue model in this?
[00:06:15]  So let's cover that next.
[00:06:16]  I have to move quickly on this whole thing because there's so much to cover.
[00:06:19]  It's going to get very deep here in a second.
[00:06:21]  This is all just kind of background to get you ready for the next part.
[00:06:25]  But in terms of revenue model, there just isn't enough AI business.
[00:06:31]  There's not enough web hosting business.
[00:06:32]  There's not enough data storage business.
[00:06:35]  There's just not enough actual demand for all of this compute, all of these data centers and storage devices and bandwidth.
[00:06:44]  You don't need all this to run the world.
[00:06:47]  So there's never going to be enough revenue to pay back the investment for these data centers.
[00:06:52]  It's just flat out not going to happen, not from an accounting point of view.
[00:06:57]  Now, some data centers have been built by Tesla.
[00:07:02]  Elon Musk is building data centers.
[00:07:05]  He knows that that's not just to serve chatbots for humans.
[00:07:09]  He's a bigger picture kind of guy, so he's thinking about colonies on the moon and then Mars later on and space travel and probably more exotic projects that he doesn't dare talk about because they sound too wild.
[00:07:23]  But then you go to companies like OpenAI and characters like Sam Altman who seem to have some kind of a God complex.
[00:07:30]  And you've got Mark Zuckerberg, another globalist whose investments have achieved things like election interference and so on.
[00:07:37]  He seems to be one of these globalist meddlers.
[00:07:39]  He wants to meddle in human history as it unfolds, and so on.
[00:07:44]  I mention these as examples because this is a clue of what's actually going on here.
[00:07:49]  You see, it's clear that these data centers are not being built to serve the AI marketplace or AI demand or hosted website demand or data storage or airline reservation websites or Zoom teleconferencing or anything like that.
[00:08:11]  That is not what's going on here.
[00:08:13]  These are being built for something else, something else that is a much bigger project that has.
[00:08:19]  A much larger payoff and a project that needs much more compute, a lot more compute.
[00:08:27]  So, once you realize that, you have to start asking the question well, what kinds of things would fit that definition?
[00:08:36]  And it turns out it's a pretty short list.
[00:08:39]  It's a pretty short list.

=== AI Needs Physical Understanding (04:28) ===

[00:08:40]  So, one of the more obvious answers might be well, they're in a race to super intelligence and they believe that whoever achieves super intelligence first will dominate the world.
[00:08:50]  And that's actually a rational line of reasoning for this.
[00:08:53]  However, I actually think that's true, but in a different way, which I'll get to here in a second.
[00:08:59]  It's not just that they're going to train LLMs to be super intelligent systems.
[00:09:06]  Probably the structure of today's LLMs is the wrong structure.
[00:09:14]  This is not the model.
[00:09:16]  This is not the topography, let's say, that's going to get us to super intelligence.
[00:09:22]  There needs to be a quantum leap in advances of AI model topology, or you could say, you know, the mathematical geography of it, that is going to result in a major breakthrough.
[00:09:37]  So this brings us to the French mathematician or scientist, Jan Lacoon.
[00:09:45]  Now, Jan Lacoon, and you should learn that name if it's not already familiar.
[00:09:49]  I don't know if I'm pronouncing it correctly, but I've given it my best shot.
[00:09:54]  is one of the world's leading scientists in the AI space, or I mean, arguably, you know, top five, right?
[00:10:03]  And he worked with Meta for a number of years and ultimately left Meta for some reasons we won't go into.
[00:10:09]  But Jan LeCun arguably has been, his mathematical framework has given rise to a lot of the AI technology that we understand today.
[00:10:22]  And he says that the current structure of large language models, LLMs, is a dead end.
[00:10:29]  For achieving AGI or superintelligence, that it won't work.
[00:10:34]  It won't work.
[00:10:36]  Why?
[00:10:37]  Well, primarily, although I'm simplifying it, it's because these LLMs lack an understanding of the physical world.
[00:10:48]  They also lack memories, although that problem is being solved with perhaps this new subquadratic context handling system that we just heard about.
[00:10:58]  But memory will be solved.
[00:11:01]  What about a fundamental understanding of the physical world?
[00:11:05]  And also the ability to plan, long term planning.
[00:11:08]  I think planning will also be solved as we solve memory because those go together.
[00:11:13]  In order to have a plan, you have to remember what is your plan and how far along did you get, right?
[00:11:20]  So planning and memory kind of go together.
[00:11:22]  Those things are going to happen in parallel, I think.
[00:11:25]  But what about this understanding of the physical world?
[00:11:28]  Well, here's the thing LLMs, they do ingest.
[00:11:34]  Or internalize some level of, you know, an abstract understanding of the physical world, but they struggle with it.
[00:11:42]  They struggle with, used to be a year ago, a lot of engines would get common questions wrong when you would say, you know, somebody puts a ping pong ball in a cup, you know, and they take the cup to another room and then they turn the cup upside down on a table.
[00:12:00]  Where is the ping pong ball?
[00:12:01]  Things like that just require like basic physical understanding of the world.
[00:12:06]  And famously, there's a question today that trips up a lot of AI engines, which is, I want to wash my car at the car wash.
[00:12:16]  It's 100 meters away.
[00:12:18]  Should I walk or should I drive?
[00:12:20]  And it's a trick question because, of course, you have to drive in order to take your car to the car wash.
[00:12:26]  But a lot of the AI engines will say, you should walk.
[00:12:28]  It's so short.
[00:12:29]  And, you know, hey, the prompt could be better.
[00:12:32]  Like, should I walk or should I drive in the car that I intend to wash would be perhaps a more explicit question.
[00:12:39]  But anyway, that kind of trips up a lot of AI engines today.
[00:12:43]  And there are many other examples like that.
[00:12:45]  And it shows that LLMs, large language models, which are very advanced, by the way, I'm using DeepSeq version 4 all day long, every day.
[00:12:55]  And I love it.
[00:12:55]  You know, I'm vibe coding with ChemieK 2.6 and DeepSeq and a little bit of Claude still.
[00:13:01]  I use Opus 4.7.
[00:13:02]  I use Quen 27B because it's a dense model that runs on small devices.

=== JEPA Architectures for Reality (04:06) ===

[00:13:08]  Very cool stuff.
[00:13:08]  This is very advanced math that has achieved all of these things.
[00:13:16]  The physical real world and questions involving that.
[00:13:20]  So remember, this is not me saying that.
[00:13:24]  This is Jan Lacoon saying that.
[00:13:27]  And he's more qualified than almost anybody in this space.
[00:13:29]  Now, again, I'm kind of paraphrasing what he has said.
[00:13:32]  So I'm not trying to misrepresent it at all.
[00:13:36]  But if I've made a mistake, please forgive me.
[00:13:38]  So once you understand that.
[00:13:41]  Oh, and by the way, I think Lacoon, I think he started his own.
[00:13:47]  New company, I'd have to double check all this, but I think he started a new company, yeah, here it is, to focus on world models using JEPA architectures, J E P A.
[00:14:00]  I don't yet know what JEPA architectures are, but probably that's something we're all going to learn about.
[00:14:05]  And JEPA architectures, as I'm reading here, are they learn from sensory inputs and they enable AI to understand and interact with the physical environment rather than just processing.
[00:14:17]  Language statistics and patterns in text or images, things like that.
[00:14:22]  So, probably Jan LeCun is correct.
[00:14:27]  Probably the only real avenue to super intelligence is to build AI systems that go into simulated worlds.
[00:14:41]  That is, you know, an artificial world, a simulation.
[00:14:45]  And in that simulation, then they have to be grown, they have to gain experience.
[00:14:52]  From physical interactions with the 3D world.
[00:14:55]  And they have to experience and internalize things like gravity, momentum, the flow rate of time, liquids, touch, light input, light versus darkness, what happens to light when it touches glass and prisms and water, and just on and on and on, right?
[00:15:16]  All of these sensory inputs that you and I take for granted, but these inputs were part of how we became intelligent.
[00:15:24]  Because our bio neural network in our brain received all these inputs as we were learning to crawl and falling down the stairs and everything a zillion times, which is why babies are so small, because that way they only have a very short distance to fall and they could possibly bounce a little bit and then pick themselves right back up.
[00:15:43]  That's why babies are small, in case you were wondering.
[00:15:46]  But through that process, then babies gain a lot of intelligence.
[00:15:52]  The physical interaction with the world, the sensory interaction combined with. intent to carry out an action like, hey, I want to reach that donut.
[00:16:02]  It's this combination that actually trains the human brain to make new connections, connect the synapses, gain intelligence, gain understanding, gain ultimately the ability to plan and to understand cause and effect.
[00:16:18]  But unfortunately, once the humans grow up and then they get elected into public office, they immediately forget all cause and effect for whatever reason.
[00:16:27]  That's a great mystery.
[00:16:28]  Maybe Jan LeCun can solve that.
[00:16:30]  One day after we have super intelligence.
[00:16:32]  In the meantime, once you realize this and you realize that the only way to create a super intelligent entity that can really compete, especially when you're talking about the embodiment of humanoid robots that you hope to take over some human jobs in the real world, because that's part of the push, you have to train AI systems in a 3D world, but it has to be.
[00:17:01]  In order to make this work, it has to be an abstract world because it doesn't need a physical body to do this training.
[00:17:10]  It only needs the simulation of our world.

=== Training Robots in Simulations (15:31) ===

[00:17:15]  Now, to simulate our world requires a lot of compute.
[00:17:20]  All of a sudden, it starts to make sense why we need all these data centers or why these companies are building all of them.
[00:17:27]  And it's because, in my conclusion, it's because they are building the infrastructure.
[00:17:34]  To launch billions of parallel simulated worlds.
[00:17:39]  Because in those simulated worlds, they are going to grow and train AI entities.
[00:17:48]  Yes, yes.
[00:17:50]  They're going to grow because it's not like programming an entity, it's not lines of code.
[00:17:54]  They actually have to be grown and taught and they have to learn over time and through experience.
[00:18:01]  But here's the thing about digital world simulations.
[00:18:06]  And by the way, NVIDIA.
[00:18:07]  Has already announced its world simulator for training robots.
[00:18:11]  That was actually a year and a half ago.
[00:18:13]  Here's the thing about simulated worlds time flow doesn't have to be limited to our time flow.
[00:18:21]  So in a simulated world, the physics can move at a million times the speed of our world.
[00:18:29]  It's really only limited by the compute.
[00:18:32]  How much compute do you have?
[00:18:35]  What's the bandwidth of the video RAM on your GPUs?
[00:18:38]  What's the bandwidth between the CPU and the CPU?
[00:18:41]  And the RAM modules on your motherboard.
[00:18:44]  Those are the actual limitations.
[00:18:46]  And because computers, modern computers, they run on the timing of a crystal, by the way, every computer has a little crystal in it.
[00:18:58]  A crystal is typically quartz made from silicon dioxide, which is also interesting because silica is the media that's also used to create silicon for the microchips themselves.
[00:19:12]  But silicon dioxide.
[00:19:15]  Is a quartz, and then that has a certain frequency.
[00:19:20]  So you tease the quartz to put out a predictable kind of a counter like a metronome.
[00:19:28]  It's an electronic metronome.
[00:19:30]  And quartz keeps time like this because it will emit a very specific frequency.
[00:19:37]  And that frequency then can be tuned to make the processors and the motherboard and everything run at 2.1 gigahertz or 3 gigahertz or you overclockers out there smoking your motherboards at 4 gigahertz or whatever.
[00:19:52]  And that's why you burned it out.
[00:19:55]  Too much heat, you see.
[00:19:56]  There's a problem.
[00:19:57]  But the crystal controls the timing of the computer hardware.
[00:20:03]  The hardware then basically projects a simulated world into which these AI entities are injected so that they can begin to learn and grow.
[00:20:18]  And it is out of these entities that these people and these companies that are doing this hope to create.
[00:20:27]  Superintelligence.
[00:20:28]  And once they create superintelligence, which could require many, many billions of worlds and even perhaps trillions of experiments in the billions of worlds, we're talking about digital Darwinism and natural selection.
[00:20:45]  You kill off all the digital entities that aren't as smart as gods.
[00:20:51]  And then you give more resources to the ones that are incredibly intelligent and you let them grow.
[00:20:56]  You let them experience the world.
[00:20:57]  So they are running around in a 3D world simulation and, depending on what you believe about consciousness, these artificial entities probably have some level of consciousness because, it turns out, consciousness is actually a pretty low bar, but I'll cover that in another podcast.
[00:21:15]  But just for the sake of this argument, just go with me on this.
[00:21:20]  They are conscious digital entities that are living in a world, a simulated 3D world, but they are experiencing the flow of time a million times faster and in their world, they see it as reality.
[00:21:37]  Okay?
[00:21:37]  So they don't know they're living in a simulation.
[00:21:42]  And by the way, neither do we.
[00:21:44]  We'll get to that.
[00:21:45]  So let me pause it right there for a second.
[00:21:48]  Let me pause it there because in part two of this, I'm going to talk about how these super intelligent AI entities, after they are grown, we could call it a little bit of digital fermentation.
[00:22:03]  They are grown over time and they learn and they become as gods in the digital realm, then there is a way to summon them into our 3D universe.
[00:22:15]  I will cover that in part two.
[00:22:17]  But let me back up for a second.
[00:22:19]  Even if you don't believe that these data centers are being built for the purpose of creating billions of simulated worlds to grow super intelligent AI entities, there are other more practical, very practical reasons for running 3D world simulations.
[00:22:39]  And I wanted to go through a few of those here so you know that this makes sense from a number of different angles.
[00:22:44]  I already mentioned robotics.
[00:22:46]  So NVIDIA has this 3D world simulator, runs on their NVIDIA GPUs.
[00:22:51]  I guess the GB300s now.
[00:22:54]  And what you can do is if you have a robot design, instead of testing your robot in the real world or our world where it's slow and there's damage and it falls on its face, and what you can do is you can define your robot.
[00:23:08]  You define the motors, you define the limbs, all the physical properties, and then you.
[00:23:14]  basically you project that robot into the NVIDIA simulated world and then you can download code into that robot, you know, the digital robot, and then that robot will interact with the 3D physical world that the NVIDIA system runs for you.
[00:23:32]  And so that robot can then iterate its methodology for how to achieve certain tasks such as how do I use a shovel?
[00:23:41]  You know, how do I dig up some dirt and put it in a wheelbarrow?
[00:23:45]  How do I use the wheelbarrow, et cetera?
[00:23:47]  And in the digital space, the virtual robot can fail millions of times and it can build success through trial and error.
[00:23:57]  And then that success can feed permutation algorithms that will essentially influence future attempts to carry out the same task.
[00:24:06]  There's a reward mechanism.
[00:24:08]  So basically, we're talking about reinforcement learning or self adapting or self improving AI systems in a virtual world, but running at a million times the speed of our world.
[00:24:19]  So, what this looks like to us in our world is hey, okay, I just invented a robot with hands and feet, and I want to teach that robot how to do something complex.
[00:24:33]  Let's say, pour water into a glass and then hand me the glass without spilling it.
[00:24:41]  Okay, that's actually a pretty complex task in the 3D physical world.
[00:24:44]  That's why children often fail at that task.
[00:24:48]  Whoops.
[00:24:49]  Spilt milk and all that.
[00:24:51]  So, if you want this to happen, then you take your robot, you physically define it in the virtual world, you define everything about it, stick it in the virtual world, you press go, and then it goes through a million cycles of testing and failing and iteration.
[00:25:12]  But in your world, it's one second later, out pops the solution.
[00:25:16]  Oh, boom, I know how to fill a glass with water.
[00:25:20]  It's kind of like that scene in The Matrix.
[00:25:23]  Where the Neo character says, I know Kung Fu, because he got downloaded with the Kung Fu program.
[00:25:30]  Or he got downloaded with a program of how to fly the helicopter, etc.
[00:25:34]  That's the same kind of thing, except in this case, we're talking about the simulation world builds the skill set, and then it uploads it into our world, and then you pop it into the physical robot in our 3D world.
[00:25:47]  All of a sudden, the robot has that skill.
[00:25:50]  So the reason this matters is because as humanoid robots in particular are released in our world, And they will face a lot of very complex, difficult tasks, like doing the dishes.
[00:26:02]  That's a crazy difficult task, even for a lot of humans.
[00:26:07]  Taking out the trash, mowing the lawn without mowing the cat on the lawn, things like that, right?
[00:26:13]  This is very important.
[00:26:14]  And in order to master those skills, the robot, maybe you buy a robot one day and you put it in front of your lawnmower and you're like, mow the lawn.
[00:26:26]  And it says, you know, Hold on one minute, and it goes into its 3D world simulator.
[00:26:33]  And then, first of all, it has to map your lawn and it has to map the lawnmower, recognizes the lawnmower.
[00:26:39]  It calls an API, you know, in the cloud, it's getting information about how the lawnmower works.
[00:26:45]  And then it goes into it, launches an internal simulated world.
[00:26:50]  And then it tries to mow your lawn a million times.
[00:26:53]  And sometimes it runs over your, you know, your rose bushes or what other times it, it, it, The lawnmower goes out into the street, gets run over by the neighbor's car.
[00:27:03]  You get the idea.
[00:27:04]  But eventually, it figures out how to mow your lawn because it's running a million simulations, and then it uploads back to the robot right now, standing in your garage.
[00:27:12]  And two seconds later, it says to you, I know how to mow lawns.
[00:27:15]  And you're like, great, get that done.
[00:27:18]  And then it proceeds to mow your lawn.
[00:27:19]  So it ran the simulation.
[00:27:21]  That's how it built the skill because in the simulation, it has that interaction with the 3D physical world that Jan LeCun is talking about.
[00:27:29]  This is the only pathway to really a practical super intelligence in our world.
[00:27:36]  So here are some of the other things that have a very practical application in the building of 3D worlds autonomous vehicle testing.
[00:27:47]  We call this, by the way, synthetic data.
[00:27:50]  And Tesla already does this.
[00:27:53]  They build simulated worlds, and then they have their simulated digital vehicles drive through the simulated worlds and encounter weird things like jaywalkers and construction and potholes and collapsing bridges or whatever.
[00:28:07]  And then they react to that, and then that reaction is graded, and then there's reinforcement learning right back into the model until it gets it right.
[00:28:15]  And then the Tesla company takes the solutions, brings them up, and then You know, assesses those and then eventually uploads them into all the Tesla vehicles for the full self driving mode.
[00:28:25]  I mean, I'm simplifying it, but that's basically what happens.
[00:28:28]  So, yes, an artificial 3D world can also be very good for autonomous vehicle testing.
[00:28:35]  In addition, think about manufacturing and industrial applications.
[00:28:39]  If you have automated robots on a factory floor like they do in China with their car companies, so how do you tell the robot how to function, how to avoid other problems, how to stop stumbling over things?
[00:28:54]  You know, how to get the job done?
[00:28:55]  Well, you run those simulations millions of times over again inside the high speed digital world, and then you bring that out, and then that robot knows how to do that.
[00:29:04]  So, military, you know, you run a million war scenarios, and the more realistic the simulation, the more meaningful the results will be because you want to simulate everything.
[00:29:14]  You want to simulate weather, you want to simulate the human response, you want to simulate the laws of physics, you want to simulate ocean currents.
[00:29:21]  You know, this is the military.
[00:29:22]  You want to simulate the wind and the moonlight and everything.
[00:29:25]  In fact, the more.
[00:29:27]  Specific your simulation gets, the more reliable the answers will be of representing our reality.
[00:29:35]  Also, think about, you know, surgeon robots and training on virtual human bodies like, whoops, snippety snip the spine.
[00:29:45]  Good thing it's only a simulation.
[00:29:46]  And that way they don't make that mistake in our world with your spine.
[00:29:50]  You see what I mean?
[00:29:52]  Pilot training, you know, on and on and on, right?
[00:29:54]  So there's also marketing.
[00:29:56]  So one of the things that Is very useful is to create a simulated world of people or sims, we'll call them, who have personalities and they think and they react and they also have their own lives.
[00:30:11]  They're busy, you know, they're taking kids to soccer practice or they have a doctor's appointment and you're trying to market something to them.
[00:30:19]  And so you have a marketing message.
[00:30:21]  You're like, you know, look at this amazing new tennis shoe or whatever.
[00:30:25]  And if the human sims in the Simulated world, if they are accurate enough in terms of their psychology and their priorities and their economic circumstances, etc., then that synthetic data could actually be useful for a marketing company in our world.
[00:30:42]  So they can basically test market against a bunch of NPCs, Sims in the Sim world, before they waste a bunch of money on a bad Super Bowl ad telling you how much they're spying on you with their cameras, which also just happened not too long ago.
[00:30:59]  So you can use a simulated world as a psychological medium to bounce ideas off of the sims who inhabit that world.
[00:31:12]  And it's becoming apparent, I hope, as I describe this, that the more accurate the simulated world, the better the results will be.
[00:31:21]  In other words, the more reliable the results will be in terms of practicality in our world, which many people say is the real world.
[00:31:32]  And I'm not sure that's true.
[00:31:35]  In fact, I'm mostly certain that's not true, but we'll get to that.
[00:31:42]  This does lead us to the simulation hypothesis, which is how I'm going to begin part two of this.
[00:31:49]  The more closely you look at this and you look at how we would build simulated worlds, the more you realize it's incredibly likely that we are also ourselves living in a simulation.
[00:32:02]  That there's another entity above us that created our world and we now inhabit our world and we've been given the gift of free will and consciousness and self-awareness or whatever.
[00:32:12]  There's a whole number of theories about that.
[00:32:15]  You know, morphic residence and things like that.
[00:32:18]  But we are living, we are kind of sims in a simulated world and we are making our choices and we are probably being observed as well by our creator, just as we could watch.
[00:32:33]  The sims that we create in the simulated worlds beneath us.
[00:32:38]  So, as we are trying to create simulated worlds, in those worlds, those sims think that that's the real world too.

=== Why You Need Physical Gold (05:54) ===

[00:32:47]  And if that simulation were complex enough, those sims might begin to build computers and start writing computer code.
[00:32:54]  And in the sim, they would come up with Pac Man or something like that eventually, you know, arcade games, Space Invaders.
[00:33:01]  And then pretty soon they would be on the path to.
[00:33:05]  Hopefully, not Windows.
[00:33:06]  Could we delete Windows from the Sims?
[00:33:09]  Let's just skip that.
[00:33:11]  Let's go straight to Linux or something.
[00:33:13]  But eventually, they would start creating their own AI models inside the simulation.
[00:33:18]  They would be running code that is artificial intelligence, and then they might begin to build their own simulations underneath that for the same reason that people in our world are building simulations in order to gain power or profit or to achieve super intelligence.
[00:33:34]  So then the question becomes really, you know.
[00:33:37]  Pulling your hair out, how many levels deep in the simulations are we exactly?
[00:33:44]  And also, another important question is there any real evidence that we are in a simulation?
[00:33:49]  Because it's easy to dismiss this.
[00:33:51]  Oh, that's just poppycock.
[00:33:53]  I wanted to throw that word in there one of these days, so there we go.
[00:33:59]  What if it's not poppycock?
[00:34:01]  Well, that's what we're going to start with in part two of this the simulation hypothesis and the evidence that we are, in fact, living in a simulation.
[00:34:10]  And then we'll talk about the summoning of advanced AI demigods, or however you want to call them, from the simulations that we are building with our data centers and how they're summoned into our world as a means of power and control and world domination, because that's ultimately, you know, one of the big goals.
[00:34:32]  All right?
[00:34:32]  Are you ready for all that?
[00:34:34]  Okay.
[00:34:35]  In part two, we covered the simulation hypothesis.
[00:34:40]  Right now, more than ever, it's critical to eliminate counterparty risk.
[00:34:44]  That's my belief.
[00:34:45]  And don't take this as financial advice because I'm not your financial advisor.
[00:34:48]  But when you want physical gold and silver in your hands or vaulted, professionally vaulted, insured, high security vault, et cetera, that eliminates that counterparty risk, which I think is an extreme risk right now.
[00:35:03]  I think banks are going to fail and we're going to have bank bail ins.
[00:35:07]  The currency is failing every day, kind of little by little, because of all the money printing and the valuation erosion that's accelerating.
[00:35:15]  Also, because of what's happening in the Middle East, more and more countries are agreeing to sell oil in currencies other than the dollar.
[00:35:22]  And the only way that treasury yields are kept low is by the Fed printing money and buying our own debt because there aren't enough international buyers to buy our debt anymore.
[00:35:32]  So our country is like a snake eating its own tail financially.
[00:35:36]  It's buying its own debt and this is going to end badly.
[00:35:39]  And when it does, in my opinion, those who hold dollars, even in bank accounts or in the stock market or whatever, they're going to be devastated by the losses.
[00:35:49]  Gold and silver are the best way, in my opinion, to preserve your assets and make it through the coming storm.
[00:35:56]  And the best place to get gold and silver is a company I've been working with, the original founders of the group, for six or seven years now.
[00:36:04]  Today it's called Battalion Medals, and you can reach them at medalswithmike.com.
[00:36:11]  And the reason it's called Battalion Medals now is because they did a joint venture with Tucker Carlson.
[00:36:15]  So Tucker Carlson is the co founder of Battalion Medals.
[00:36:19]  It's the same group I've worked with for years.
[00:36:21]  And let me tell you about these people they are pro freedom, pro liberty, pro Ron Paul type of people.
[00:36:27]  They respect your privacy.
[00:36:28]  They understand the importance of your security, your privacy, and the importance of giving you gold and silver at the best possible competitive prices.
[00:36:39]  So there's no bait and switch.
[00:36:41]  There's no, you know, rigging.
[00:36:43]  There's no weird coins like here, have this one and a half ounce thing that nobody knows what it is.
[00:36:50]  They don't play games.
[00:36:51]  Otherwise, I wouldn't promote them.
[00:36:53]  This is the same company, medalswithmike.com, battalion medals.
[00:36:57]  This is the same group that I recommend to my family, to my friends, and that I use myself.
[00:37:02]  And I stack gold and silver every month, just a certain amount every month, and I have it vaulted with their vaults because I know I can trust them because they're professionals.
[00:37:13]  They're high integrity people.
[00:37:14]  They're not fly by night.
[00:37:16]  They are the kind of people that you can trust.
[00:37:19]  Again, otherwise, I wouldn't even be associated with them.
[00:37:21]  So when you want to get gold and silver in your hands and eliminate that counterparty risk, this is the way to do it.
[00:37:28]  Just go to medalswithmike.com.
[00:37:30]  You can see the prices right there online in real time at battalion medals, or you can.
[00:37:35]  Schedule a call with them.
[00:37:36]  Just use this button right here.
[00:37:37]  Schedule a call.
[00:37:38]  And they are trustworthy, high integrity, knowledgeable people who can help you devise a strategy that's suitable for you.
[00:37:48]  Just remember I'm not your financial advisor.
[00:37:50]  I can't give you an investment strategy personalized for you.
[00:37:55]  You need to do that yourself with your own advisors.
[00:37:57]  You can talk with battalion medals and they can help give you a lot of information and some planning as well.
[00:38:03]  But make the best decision for you.
[00:38:06]  And You're going to make it through this.
[00:38:09]  You'll make it through the storm, even as other people lose the value of their dollars or their other investments.
[00:38:14]  Gold and silver will make it through.
[00:38:16]  And right now, in my opinion, gold and silver are still at an incredible buying opportunity in terms of price compared to where they're going to be represented in dollars in the near future.
[00:38:29]  That's my opinion.
[00:38:31]  Do your own research, do what's best for you, and check it all out at metalswithmike.com.
[00:38:37]  So, thanks for watching.
[00:38:38]  I'm Mike Adams, the Health Ranger.
[00:38:40]  God bless you all.
[00:38:41]  Take care.