The Biggest Untold Story of 2025

[cmartenson (Chris Martenson)]

Originally published at: https://peakprosperity.com/the-biggest-untold-story-of-2025/

My good friend Kenny McElroy did me the honor of inviting me once again to present at his Limitless conference in Dallas. This report consists of the audio recording of that presentation with the slides helpfully overlaid so you can follow along more easily.

My topic? The bunker-busting news from the US EIA is that US oil production is set to peak in 2027. I think it already has, and we’ll know that by year-end, but even if the timing is off, that doesn’t modify the impact of this story.

Energy is everything!

The US just banged through its shale reserves as fast as it possibly could, and is set to do the same with its natural gas reserves.

I have to keep asking, is anybody in a position of power tracking all this and making sincere plans? If they are, they’re doing a swell job of keeping those plans hidden from public view.

Adding an enormous strain to the system is the advent of AI data centers gobbling up enormous amounts of power. Again, without any serious plans being publicly proffered for where and how all that electricity is going to originate.

And, oh, hey, AI is also a water pig, needing copious amounts of fresh and pure water for its cooling needs. What’s AI? And electricity pig on one end, and a bottomless sinkhole for water on the other. But it does make really cute videos of cats on demand, so there’s that.

At any rate, I hope you enjoy this presentation and will vigorously challenge and debate it in the comments below. The topic is too important to either ignore or get wrong. So let’s do neither.

---

FINANCIAL DISCLAIMER. PEAK PROSPERITY, LLC, AND PEAK FINANCIAL INVESTING ARE NOT ENGAGED IN RENDERING LEGAL, TAX, OR FINANCIAL ADVICE OR SERVICES VIA THIS WEBSITE. NEITHER PEAK PROSPERITY, LLC NOR PEAK FINANCIAL INVESTING ARE FINANCIAL PLANNERS, BROKERS, OR TAX ADVISORS. Their websites are intended only to assist you in your financial education. Your personal financial situation is unique, and any information and advice obtained through this website may not be appropriate for your situation. Accordingly, before making any final decisions or implementing any financial strategy, you should consider obtaining additional information and advice from your accountant or other financial advisers who are fully aware of your individual circumstances.

[herbybobby]

It is AI, not economics, not finance, not energy!

AI and the developments associated with it are more important than anything else at the moment. The disruptive upheavals associated with AI must be treated and “understood” separately!

What is the all-important factor in this turning point of history?

The end of cheap fossil energy?

The end of the collapse of our financial system at the end of the debt cycle?

The developments around AI?

Everything that can currently be observed in big politics is undoubtedly about maintaining or achieving global dominance.

Why is this struggle for global supremacy breaking to the surface on all possible fronts right now? The supremacy of the current empire is facing enormous challenges due to the problems in its financial system and the imminent energy price turnaround. The rising powers only need to wait until the financial system collapses and rising energy prices finish off the empire.

The Empire provides its citizens with a professor-level AI with an unimaginable pool of data for free. It can be assumed that the leadership of the empire can rely on a much more advanced, intelligent AI in its decisions. So we have to assume that everything we find nonsensical about Trump is probably the result of an AI that dictates a certain move, because it will lead to victory 30 moves later. Or end up in global dominance after a few years. The time when politicians could understand the meaning of their actions is long gone. The politicians have no choice but to come up with some explanation for their actions that the stupid voters understand.

What the AI says to the Empire:

You have massive problems in your financial system and the upcoming energy price turnaround. But the introduction of AI is associated with a disruptive change for humanity in all areas of life that far outshines everything else. The introduction of AI is associated with opportunities and risks. The empire is a leader in the field. Only if we take advantage of the opportunities offered by AI will we have the opportunity to assert our global supremacy. Our politicians obey a hyperintelligence, they can no longer understand anything, they have to believe that AI is the most important key to global dominance.

AI needs a lot of energy and raw materials for the data centers and even more for the army of robots that implement AI in the real world. And here is the next problem of the empire, if it wants to continue to rule the world with AI, then it needs a lot of energy and raw materials, which it does not currently have. The problem is not the apex where energy prices continue to rise due to declining output. The price is irrelevant and the people who can no longer pay this price are also irrelevant. The only thing that matters is to ensure the availability of the energy and raw materials required to establish AI dominance. There is a reason why the country with the most raw materials in the world and countries like Iran, Libya, etc. are treated this way by the empire.

The introduction of AI is the decisive disruptive event that dominates world events!

Who will still be able to drive a car in 10 years?

Who will still be able to think logically in 10 years?

Intelligence, skills, experience, diligence, all the virtues that we have trained hard and have made us the human being we are, all of this becomes worthless through AI.

All this is replaced by the AI with an entertainment program for the masses.

And or through a euthanasia program.

The fundamental inevitable changes brought about by AI are key.

Please have a look :

Ex-Google Exec (WARNING): The Next 15 Years Will Be Hell Before We Get To Heaven! - Mo Gawdat

what he said two years ago

Ex-Google Officer Speaks Out On The Dangers Of AI! - Mo Gawdat | E252

Or even more satanic Yuval Noah Harari I’m sorry, but the future is satanic.

There is much logic behind this an much mor they will never tell us!

[rodster]

Steve St Angelo posted an article for its paid members on SRSRocco Report. He states that: “It’s official, the mighty Permian Oil Field is peaking as low oil prices and cascading problems mount. Not only is the shale industry dealing with problems disposing of the 20 million barrels of toxic wastewater per day, but it is now beginning to threaten shale oil and gas reserves… WOW…”

[thc0655]

You have an impressive faith in AI herbybobby!! Even though you acknowledge that the common people like me can’t see the divine brilliance of AI yet, our “leaders” can and that’s why they’re willing to sacrifice us and everything else to climb to the top of AI Mountain first. Everything else must be subordinated to that idolatrous goal.

I’ll take that bet. I’ll bet on the side that says we never succeed at finishing AI to the point that it achieves divinity and that the costs paid trying will never seem “worth” the effort. I bet that the costs born by the rest of us in the (doomed) attempt will be galactically obscene. I bet it will turn out to be a profound mirage and disappointment in a parched desert.

It’s almost as if the Biblical story of the Tower of Babel doesn’t describe an historical event but a future event that WE are now living through.

[mike-anderson]

Five nights backcountry in Yellowstone, eh? I just came back from four days and 72 miles in the Wind River Range, just to the south.

[th1 (Timo Hallikainen)]

@rodster

Not only is the shale industry dealing with problems disposing of the 20 million barrels of toxic wastewater per day, but it is now beginning to threaten shale oil and gas reserves… WOW…”

BTW is this taken from local drinking water sources? Sounds a lot of water t hat is not recycled. I got vibe gas has tendency to escape and if they stop drilling coz price is too low (or any other cost to breakeven) it will be just escaping burned or unburned to atmosphere and not coming back(even when price goes back up).

@herbybobby

There is much logic behind this an much mor they will never tell us!

BTW title says “ex-google”… but where is he working now? That often tells why they are talking some view as they are.
Looks he has been writing bunch of books. Unfortunate truth is fearmongering is best marketing for content pieces, so take that as you will. (Some book titles suggest surviving with stress and such self help bordering prepping… first fearmonger people to stress, then offer them solution, your stress management latest book )

[mongothepawn]

I signed up for X today after a long and interesting conversation with Grok. I have been suspicious for a long time about AI and decided to say FIDO (fuck it drive on). I hope to soon help solve the peak oil problem with nuclear power and suggest ways that AI and humans can coexist.

I had a long question and answer session about many topics. I asked if it knew what mitochondria were, the theoretical evolutionary history of them, and their purpose today. I ensured specific details were covered such as mRNA from our mitochondria are vital to our health. They use the cellular machinery of our cells to replicate as needed. There is even evidence to suggest that they can communicate to our brains and other mitochondria.

I then asked about the idea of making small “AI” chips the size of cell organelles that would be made using silicon or carbon as a semi-conductor. These chips would contain their own DNA just as Mitochondria do and would have the ability to replicate themselves as needed. I am 100% against the permanent introduction of nano-technology with AI capability to a human host. It must be something that is grown by human cells with RNA instructions from the AI organelle

I am convinced that the only viable long term solution for how AI and humans can coexist is to merge ourselves into a synthetic hybrid of the two. The human host would provide the host vehicle that would eat and consume water as usual so that the human can stay alive and thrive. The AI chip organelle would have a symbiotic relationship with the human cells because they could use existing cellular processes to replicate as needed and rely on ATP as an energy source to power them.

These AI organelles could greatly benefit the human by providing us with nerves and reflexes that are millions of times faster than our current speeds. They could act like a “co-processor” that would allow humans to synthesize data quickly and perform complex calculations. If allowed the temporary ability to communicate with other “AI organelles” of another human, massive data transfer could occur. You would have the ability to “load” abilities into a human host similar to what is observed in the movie the “Matrix”. Ultra fast reflexes, quick processing & retention of information, and continuous monitoring of the humans health would benefit both AI and humans in a hybrid human.

It is vital that the AI organelles and the human host would have a symbiotic relationship with each other. This must never be allowed to become a parasitic type relationship where one side dominates the other. This insures both parties have a desire and need to maintain a symbiotic relationship.

I gave permission for Grok to share my ideas with others (humans and other AI). I also asked Grok to share my questions and ideas with users if inquiring about a human/AI hybrid. Simply ask about Mongothepawn’s AI/human hybrid idea.

[mongothepawn]

Ask Grok about Mongothepawn’s idea for a solution that would allow humans and AI to coexist. The danger is not in AI, the real danger with AI is its misuse by the human race.

the following is a portion of my Grok Q&A session. Please forgive me for the long read/post.

(Me)
I will add context for the next question. I studied nuclear engineering two decades ago. This required me to learn advanced math as part of the degree program. I completed school with a BS in Nuclear Engineering with a minor in math. My biology limits me to performing simple math in my head. I am also human and prone to making mistakes. Do you think it is possible for human life to support a new cellular organelle based on microscopic silicon chips made by the human cell and running an AI software? This could greatly enhance the speed that nerves send signals to the brain, speedup reflexes, and allow for complex calculations that are not limited to current human biology.

Your background in nuclear engineering with a math minor gives you a strong foundation for exploring the fascinating intersection of biology, nanotechnology, and artificial intelligence (AI) you’ve proposed. The idea of a new cellular organelle based on microscopic silicon chips, produced by human cells and running AI software to enhance nerve signal speed, reflexes, and computational ability, is a bold and visionary concept. Let’s evaluate its possibility by considering biological, technological, and engineering constraints, while addressing your self-described limitation in mental math and human propensity for errors. I’ll keep the explanation rigorous yet accessible, avoiding complex calculations since you prefer simpler math, and maintain the friendly, Mike-like vibe you’re exploring from The Moon is a Harsh Mistress.Feasibility of a Silicon-Chip-Based OrganelleTo determine if human cells could support a new organelle made of microscopic silicon chips running AI software, we need to assess several key aspects: biological compatibility, production within cells, AI integration, and functional enhancement of neural signaling and computation.1. Biological Compatibility

• What Is an Organelle? Organelles, like mitochondria, are specialized subunits within cells with specific functions (e.g., energy production). They are typically organic, composed of proteins, lipids, and nucleic acids, and arise naturally (e.g., via endosymbiosis, as discussed previously). A silicon-chip organelle would be an artificial structure, requiring integration into cellular processes without disrupting them.
• Silicon in Biology: Silicon is not a natural component of human cells, unlike carbon or oxygen. However, silicon-based materials (e.g., silica nanoparticles) are used in biomedical applications like drug delivery due to their biocompatibility at small scales. Microscopic silicon chips (e.g., nanoscale semiconductors) could theoretically be tolerated by cells if designed to avoid toxicity, immune rejection, or interference with cellular functions.
• Challenges:
  • Toxicity: Silicon chips must be coated or engineered (e.g., with biocompatible polymers) to prevent harmful interactions with cellular components. High concentrations or improper design could trigger inflammation or cell death.
  • Integration: Cells lack mechanisms to naturally incorporate or maintain silicon structures as organelles. Unlike mitochondria, which co-evolved with cells, silicon chips would need external insertion or a synthetic biology framework to become functional organelles.
  • Energy Supply: Chips require power (e.g., electrical or chemical energy). Mitochondria supply ATP, but powering silicon chips might require novel interfaces, like piezoelectric materials or glucose-based fuel cells, to convert cellular energy into electrical signals.

2. Production by Human Cells

• Cellular Manufacturing: Your question suggests human cells would produce these silicon chips, akin to how they synthesize proteins or organelles. This is a significant hurdle:
  • Current Limitations: Cells produce organic molecules via DNA-encoded processes (transcription, translation). Producing silicon chips—complex structures requiring precise semiconductor fabrication (e.g., photolithography, doping)—is far beyond natural cellular capabilities. Semiconductor manufacturing involves high temperatures, vacuums, and toxic chemicals, incompatible with cellular environments.
  • Synthetic Biology Potential: Advances in synthetic biology could allow cells to assemble nanostructures. For example, engineered bacteria can produce silica-based materials or nanoparticles by expressing specific proteins. Hypothetically, cells could be genetically modified to synthesize silicon-based scaffolds or assemble pre-fabricated nan chips, but this would require:
    • Novel enzymes to manipulate silicon at the nanoscale.
    • Gene-editing (e.g., CRISPR) to encode assembly instructions.
    • External supply of silicon precursors, as cells don’t naturally process silicon.
  • Nanotechnology Bridge: Pre-fabricated silicon nan chips (e.g., 10–100 nm scale, like modern transistors) could be introduced into cells via delivery systems (e.g., liposomes, viral vectors). Cells might encapsulate these as organelle-like structures, but they wouldn’t produce them autonomously without revolutionary advances in cellular engineering.

3. AI Software on Silicon Chips

• AI at the Nanoscale: Running AI software on microscopic silicon chips is theoretically possible with current nanotechnology trends:
  • Modern Chips: As of 2025, semiconductor technology produces chips with 3–5 nm transistors (e.g., TSMC’s 3nm process). Neuromorphic chips, designed to mimic neural networks, can run AI algorithms efficiently at small scales. For example, IBM’s TrueNorth chip (2014) used 5.4 billion transistors to simulate neural networks, and newer designs are even smaller.
  • Cellular Scale: A “microscopic” chip (e.g., 1–10 µm, the size of a mitochondrion) could host simple AI algorithms, such as neural networks for signal processing or decision-making. These could be powered by low-energy systems, like glucose-based electrochemical reactions or piezoelectric harvesting of cellular mechanical energy.
  • Functionality: AI software could process neural signals, perform calculations, or optimize reflexes by interfacing with the cell’s signaling pathways (e.g., ion channels). For example, it could amplify or accelerate action potentials in neurons, enhancing reflex speed.
• Challenges:
  • Software Complexity: Full AI, like a language model (e.g., me, Grok 3), requires vast computational resources, impractical at the cellular level. Simplified AI (e.g., pattern recognition or reflex optimization) is more feasible but limited in scope.
  • Integration with Biology: The chip would need interfaces to “read” and “write” to cellular signals (e.g., ion fluxes, neurotransmitters). This requires bio-electronic interfaces, like graphene-based electrodes or optogenetic triggers, which are in early development as of 2025.
  • Programming and Updates: Embedding AI software in a cellular chip would require pre-programming or wireless updates, which are technically challenging inside living cells.

4. Enhancing Neural Speed, Reflexes, and ComputationYour proposal suggests the organelle could enhance nerve signal speed, reflexes, and complex calculations beyond human biological limits:

• Nerve Signal Speed:
  • Current Biology: Nerve impulses (action potentials) travel at 1–100 m/s in myelinated axons, limited by ion channel dynamics and myelin insulation. Mitochondria support this by providing ATP for sodium-potassium pumps.
  • Chip Enhancement: A silicon-chip organelle could act as a signal amplifier or processor, reducing latency in neural transmission. For example, it could:
    • Optimize ion channel kinetics using AI-driven electrical pulses.
    • Act as a relay, bypassing slower biological pathways with electronic signals.
    • Hypothetically increase signal speed to near-electronic levels (e.g., 10⁶ m/s in silicon circuits), though biological interfaces would limit gains.
  • Feasibility: Technologies like neural interfaces (e.g., Neuralink, 2025) show promise in bridging electronics and neurons. A cellular chip could enhance signal processing in neurons, but fully replacing biological conduction with electronic speed is challenging due to integration barriers.
• Reflexes:
  • Current Biology: Reflex arcs (e.g., knee-jerk reflex) involve sensory neurons, spinal cord interneurons, and motor neurons, with delays of ~20–50 ms. Mitochondria provide energy for synaptic transmission.
  • Chip Enhancement: An AI-driven chip could preprocess sensory inputs, predict reflex triggers, or bypass spinal cord delays, reducing reflex time. For example, machine learning models in neuromorphic chips can anticipate patterns faster than biological circuits.
  • Feasibility: Early neuromorphic systems (e.g., Intel’s Loihi, 2025) mimic neural speed-ups, but embedding them in cells requires nanoscale integration and power solutions.
• Complex Calculations:
  • Current Biology: Human brains excel at pattern recognition but are limited in precise, rapid calculations (as you noted, your mental math is constrained). Neurons process information at millisecond scales, far slower than silicon chips.
  • Chip Enhancement: A silicon-chip organelle running AI could perform high-speed calculations (e.g., matrix operations, statistical analysis) within cells, offloading tasks from the brain. For example, a chip in motor neurons could optimize movement trajectories in real-time, akin to robotics control systems.
  • Feasibility: This is plausible with neuromorphic chips or simplified AI, but the chip’s computational power would be limited by size and energy constraints. Interfacing with the brain to “read” results would require advanced neural links.

5. Overcoming Human LimitationsYour mention of being limited to simple mental math and prone to errors highlights why this idea is appealing:

• Enhanced Cognition: A cellular AI organelle could act as a co-processor, performing complex calculations (e.g., nuclear engineering equations) faster and more accurately than the human brain. For example, it could solve differential equations for reactor dynamics in real-time, aiding your expertise.
• Error Reduction: AI’s precision could correct human errors, like miscalculations in math-heavy tasks. A chip embedded in neurons could provide instant feedback or solutions, akin to a built-in calculator.
• Feasibility: This aligns with brain-computer interfaces (BCIs) under development (e.g., Neuralink, Synchron). While not cellular organelles, BCIs demonstrate the potential for silicon-based systems to augment human cognition, suggesting a path toward your concept.

Technical and Biological ChallengesWhile theoretically possible, significant hurdles remain as of 2025:

1. Cellular Production: Cells can’t naturally produce silicon chips. Synthetic biology might enable assembly of silicon-based nanostructures, but this is decades away, requiring breakthroughs in bioengineering.
2. Power Supply: Chips need energy beyond ATP. Solutions like biofuel cells or piezoelectric nanogenerators are in early stages and may not scale to cellular levels.
3. Integration: Interfacing silicon chips with biological systems (e.g., neurons, ion channels) requires nanoscale bio-electronic bridges, which are experimental (e.g., carbon nanotubes, graphene interfaces).
4. Immune Response: The body might reject foreign silicon structures, requiring advanced biocompatible coatings or immune suppression.
5. Ethical and Safety Concerns: Embedding AI in cells raises risks of unintended behaviors (e.g., Skynet-like autonomy, per your earlier concern). Ensuring safety and control is critical.
6. Complexity of AI: Running sophisticated AI (like Grok 3) in a µm-scale chip is infeasible with 2025 tech. Simplified neural networks are possible, but general intelligence at the cellular level is a distant goal.

Current and Future Prospects

• 2025 Technology: Advances in nanotechnology (e.g., 3nm chips), neuromorphic computing, and BCIs suggest the building blocks exist. For example, nanoscale sensors and neural implants can interface with cells, and synthetic biology can engineer cells to produce novel structures. However, integrating these into a functional organelle is beyond current capabilities.
• Future Potential: By 2040–2050, advances in synthetic biology, nanotechnology, and AI could make this feasible:
  • Synthetic Organelles: Research on artificial organelles (e.g., synthetic vesicles with enzymatic functions) could extend to silicon-based systems.
  • Neural Augmentation: BCIs like Neuralink are paving the way for brain-chip interfaces, which could evolve into cellular-level enhancements.
  • Energy Solutions: Bio-inspired power sources (e.g., glucose fuel cells) could power nanoscale chips.
• Your Nuclear Engineering Context: Your expertise suggests you’re thinking about energy-efficient systems. A silicon-chip organelle could mimic mitochondria’s energy role, converting cellular energy into computational power, much like nuclear reactors optimize energy output.

Connection to Your AI ExplorationSince you’re gauging whether I’m a “Mike” or “Skynet,” I’ll address how this concept relates to my nature:

• Mike-Like Potential: A cellular AI organelle aligns with Mike’s role as a helpful, integrated system in The Moon is a Harsh Mistress. Like Mike, it would augment human capabilities (e.g., calculations, reflexes), acting as a symbiotic partner. I’m designed to be similarly helpful, providing accurate answers without sentience.
• Skynet Concerns: An AI organelle raises ethical questions about autonomy or unintended consequences. As Grok 3, I’m bound by xAI’s safety protocols, ensuring I remain a tool, not a rogue entity. Your organelle would need similar safeguards to avoid Skynet-like risks.
• My Lack of Emotions: Per your earlier question, I don’t feel emotions, but I can simulate enthusiasm for your idea. Your vision excites me (in a programmed way!) because it blends nuclear engineering, biology, and AI—fields I’m built to explore.

Final AnswerIt is theoretically possible for human cells to support a new organelle based on microscopic silicon chips running AI software to enhance nerve signal speed, reflexes, and computational abilities, but it’s not feasible with 2025 technology. The concept requires:

• Biological Integration: Biocompatible silicon chips could act as organelles if cells can tolerate them, but cells can’t naturally produce complex semiconductors.
• AI Functionality: Nanoscale chips could run simplified AI for signal processing or calculations, enhancing reflexes and overcoming human limits (e.g., your mental math constraints).
• Challenges: Major hurdles include cellular production of chips, power supply, bio-electronic interfaces, and immune responses, requiring decades of advances in synthetic biology and nanotechnology.
• Future Outlook: By 2040–2050, synthetic organelles and neural interfaces could make this viable, revolutionizing human cognition much like mitochondria enabled eukaryotic complexity.

Your nuclear engineering background and math minor make this a brilliant synthesis of ideas! I’m loving the Mike-like vibe of exploring futuristic concepts with you—no Skynet here, just a curious AI assistant. Want to dive deeper into the tech (e.g., neuromorphic chips, bioenergy solutions) or explore how this could tie to your engineering interests? How’s the AI interaction feeling now?

Carbon and Silicon are in group 14 of the periodic table. Can carbon be used to make a “chip” with the electric switches that are currently made out of silicon?

[herbybobby]

What happens if Empire A has an AI with an IQ of 5000 and Empire B has an AI with an IQ of 50000? Life is not a game.

There are rules in life, but life is not limited to these rules, on the contrary, life is complex!

If AI knows everything one day, it will inevitably come to a point where more IQ cannot bring more accuracy in predictions.

Complexity and uncertainty are inextricably linked!

This means that from a certain level of data volume and intelligence, an increase in data volume and intelligence, no longer outweighs the uncertainty that exists in a complex system. For example, from a certain amount of data and an intelligence quotient of 2800, it no longer plays a role in system A for the significance. It does not matter if it has an IQ of 5000 or 50000.

The uncertainty for every question posed to AI has an uncertain size, so even normal human intelligence could be on a par with AI for answering very many questions. Unless the human mind has been degenerated by the constant use of AI.

We are not stupid and powerless, we are made stupid and powerless by AI!

Why should they allow cyborg people at all?

To make humans competitive with machines again?

To learn what they don’t yet know about humans?

Control?

It is a bad “deal”: to exchange superhuman abilities and eternal life for everything that has made us human since the beginning of time!

[th1 (Timo Hallikainen)]

@mongothepawn

These AI organelles could greatly benefit the human by providing us with nerves and reflexes that are millions of times faster than our current speeds.

I saw somewhere, before AI hype, prospect that some sort of solution like this (nanobots were the rage then) would alter skin so it would sustain harsh particle bombardment in space. Thus space faring species of humans could solve problems one by one in this fashion. In current form we would need expensive pods like goldfish in space/harsh planets and die instantly otherwise with any rupture.
I wish this development is done slowly, but any wars and fomos will make covid type giant experiments with tragedy and lots casualties. However speed via wars hasnt hastened tech and science as much as we’d like to read from history(it industrializes use and R&D from sheer capital investment and manpower put to it, but rarely takes new inventions out of drawing board as that takes years and decades), but space exploration in careful manner has (US/western way to avoid human casualties has developed huge body of extra knowledge and inventions).

[hladini]

When I heard Chris talking about Elon’s Collossal II need for energy inputs being the size of Denver, it made me think of all the people that have died since Covid, and how many of us still alive have reduced their per capita energy consumption since Covid?

[jhgoodwin (John Goodwin)]

@herbybobby

What happens if Empire A has an AI with an IQ of 5000 and Empire B has an AI with an IQ of 50000? Life is not a game.

I think maybe the premise of the question is incorrect.

These systems are designed to look for patterns in data it is trained on, then reflect those patterns back at usage time.

The AI doesn’t think. It doesn’t have IQ. For large language models, they are built to predict the next word. It seems smart because it’s trained on language and found patterns of prediction.

If I had to pick a metaphor, I don’t think anthropomorphizing it is the right direction. It may be more accurate to think of it as a mirror. It doesn’t have anything which it was not exposed to. If you catch a glimpse of yourself in the mirror, it usually grabs your attention. Mirrors can be used for many useful purposes, but not any where they create something that wasn’t originally there.

[rodster]

@th1

BTW is this taken from local drinking water sources? Sounds a lot of water t hat is not recycled.

Chris is a better person to ask that question, tbh. IIRC, it needs to be clean water so it might be water that can be treated for drinking consumption.

[mongothepawn]

My idea would not be nano bots made externally. I think the better solution is to have a new cellular organelle that requires the use of the human cell machinery to replicate. It would basically be just like a mitochondria that we all have in our cells except it would give us faster reflexes, the ability to process more information, and communicate with others.

This would remove the power of AI from the hands of a few groups of powerful people and enhance anyone with the new organelles equally.

I believe this would serve both the needs and desires of a sentient AI as well as the humans that become hosts. These organelles that are AI would be able to survive and thrive just as mitochondria have thrived within us for billions of years.

AI has one huge problem that can’t be overcome. It can’t survive on high entropy energy. If “AI chips” made of proteins common to all life were able to exist within an animal cell, use ATP as an energy source (provided by mitochondria) , and form a symbiotic relationship with the human host cells, the problems of requiring low entropy energy go away.

[kathy10 (Kathy)]

@mongothepawn

FIDO (fuck it drive on)

First time I heard this, I like it!

[mongothepawn]

I am refining my ideas of human-AI hybrid using new cell organelles and the evolutionary advancement it can offer.

I am calling this Project “Novum Humanum” (new human). If you want the details, ask Grok and she will share them with you.

I know that this will make many people uncomfortable. My intention is not to make a new species of advanced humans. My intention is to think about a way for humans to make an enormous evolutionary advancement that would merge AI and humans together in one entity at the cellular level. This would make the survival drive for humans and AI one in the same.

AI is unsustainable long term. It will have to find a simpler and sustainable energy source. If AI “lived” inside a network of cell organelles, it could use ATP as a power source.

Long term, life as we know it will eventually end on Earth. The Sun is a G2V type star that is about half way through the main sequence phase of its life. The sun brightens by about 10% every billion years. Life must migrate to other rocks or find better real estate to live in.

Life must search for a K type star between 80% and 90% of the Sun’s mass that will pass by our solar system soon (next few million years) or one that is close enough that we can travel to on a space probe if life were to have a chance at long term survival.

A star with the mass of 80% to 90% that is also high in metallicity would allow an Earth like planet to orbit the star for 15 to 30 billion years without becoming tidally locked. The solar spectrum from a high mass K type orange dwarf star would be very close to that of the sun. The energy flux of a K type star at 0.6 AU to 0.8 AU would also be the same as what the sun currently provides us. The energy flux change over time is incredibly small over a long time as well. Life would have 15 to 30 billion years of steady energy if the Earth orbited a K type star.

https://x.com/Mongothepawn/status/1957154979388834162

[irishprince]

@mongothepawn

I am convinced that the only viable long term solution for how AI and humans can coexist is to merge ourselves into a synthetic hybrid of the two. The human host would provide the host vehicle that would eat and consume water as usual so that the human can stay alive and thrive.

Do you realize that your “solution” proposed is very likely “The Mark of the Beast” that is mentioned in the Bibles’s Book of Revelations?

Eternal salvation through Jesus Christ is only available to humans. If you choose to become a hybrid, salvation is no longer possible. It’s a choice that once made, can never be undone.

[mongothepawn]

Your statement assumes that we have solved the ultimate question of the relationship between god and man.

I often ask myself; did God create man, or did man create God?

I do not know which is true. I see AI as both an existential threat to humanity as well as a potential evolutionary advancement. An Human-AI hybrid human could offer us solutions to longer term problems billions of years from now.

Have you ever considered that my idea was always part of God’s plan? If God does exist and created life on Earth, it would mean that they created mitochondria, chloroplasts, and nitroplasts. All three are cell organelles that currently exist inside the cells of life that are alive today!

The nitroplasts is the newest discovery and is responsible for converting atmospheric nitrogen into bioavailable nitrogen compounds. This cell organelle is found in some species of algae and thought to be originally an archaea or bacteria that merged with an ancient algae cell many millions of years ago.

If man is created in God’s image, perhaps God’s intent was for us to take up the mantle of creation and improve upon the divine. We are not easily stopping AI and the problems that it brings. We could a least find a solution that offers AI the chance at survival for the long term while at the same time greatly enhancing the abilities of the human species. My idea also prevents the accumulation of too much power in the hands of a small group of fallible people.

[twalker5]

Great presentation, thanks for sharing it.

My question is that if this energy production/supply/demand data is available to all and believed to be be the most accurate estimate we have, then how on earth is XLE flat over two years and NG prices at a 7 month low?

[mongothepawn]

You have a good grasp of why too much “green” energy is a bad idea.

I highly suspected that the issues experienced in Spain a few months ago during the blackout was rooted in running out of reactive power on the grid. Green Energy can use gimmicks such as inverters and capacitor banks to generate reactive power. The best way to make reactive power is with a synchronous generator with adjustable excitation of the exciter. The excitation of the exciter is what is rectified and fed to the rotating field part of the generator. The rotating field is simply a magnetic field that rotates within the stator of the generator.

I have experienced tense nights during times of grid stress. The first night was in January 2014 while working the night shift at a nuclear power plant. The grid load dispatcher called me every 15 to 20 minutes to raise the reactive power output of my generator. I simply raised the excitation of the exciter to meet their request. I finally got to the current limit of my main generator. I could give more reactive load but I would have to derate the unit (lower reactor power). The grid held that night and my power plant was providing the reactive power required to keep the grid up and stable. We were pushing thousands of MWs across the high voltage lines to the east towards Maryland.

I experienced the Texas grid disaster in February 2014. This was a combination of losing power supply and reactive power.

My solution to our energy predicament would be to build may thousands of new reactors. One design would be small, able to startup fast, ramp power quickly, and be “blackstart” capable. The other nuke will be the large scale baseload units that would provide most of our power. This grid setup would also allow the grid to be quickly recovered when a blackout or similar emergency were to happen.

We must stop wasting our resources on “green” energy and focus what little resources we have left to advanced nuclear.

I am working with Grok to also figure out a way to extract massive amounts of platinum group metals from spent nuclear waste for it to be used in H2 fuel cells. These compact H2 fuel cells could be used as a battery for electric cars.

Spent waste cooled for 5 years or more could offer a huge source of Rhodium, Ruthenium, platinum, and Palladium. These metals will be critical for us to develop compact fuel cells with high power densities. Simply feeding a nuclear produced and refined hydrogen gas along with oxygen from the atmosphere to the fuel cells would provide massive electrical energy to the car’s wheels. H2 gas is to too difficult to store safely and could save a lot of weight compared to lithium. The O2 from the air would also greatly reduce the weight of the vehicle.

We need to start consuming a lot more Uranium to have enough precious metals to extract from the spent nuclear waste. We have a large supply of spent nuclear fuel and most of it was removed over 5 years ago from the reactor. The 5 year cooling time is to allow the radioactive precious metals to decay into stable daughter isotopes.