A Lesson for the Frontier AI Labs - the Process is the Key

In the mid-1980s, the Quality Revolution hit American Management like a sledgehammer. Up until then, American Management had enjoyed the benefits of the post-war fact that nearly all of the manufacturing nations of the world had essentially destroyed themselves during World War II, leaving the United States of America as the sole superpower with a powerful manufacturing capability. Yes, the Soviet Union was another superpower to contend with, but the only thing that the Soviet Union excelled at was the manufacture of 30,000 nuclear weapons and the means to deliver them. Russian dishwashers were not much of a threat on the world market. But then the recently defeated German and Japanese economies reemerged, and especially the Japanese. The Japanese manufacturing companies had adopted the Quality teachings of W. Edwards Deming, which sought the continuous improvement of manufacturing processes by means of statistical analysis. Suddenly, Americans could buy Japanese cars and cameras that actually worked without the fear of getting stuck with a "lemon". This Japanese Quality Revolution threw American Management into one of those panics that all Managements are subject to. Suddenly, all American workers had to learn about Quality and start doing it immediately or else!

At the time, I was in the IT Department of the American Amoco Oil Company. I was called into the office of my supervisor only to learn that I would be spending the next 6 months teaching a Quality Class to all members of the IT Department, including the staff members, Management, and all the Vice Presidents - a one-hour mandatory Quality Class one day each week for 8 weeks! Amoco had decided to teach the Quality Theory of Philip Crosby instead of Deming's because that avoided the statistical mathmatics found in Deming's approach. I was sent to a one-week training class at a local Crosby Institute facility to learn how to teach Quality. The textbook that I would be teaching from was a thin hardcover book that cost $125 each in 1985 dollars, and Amoco was buying 1200 copies! The book was written at an 8th grade reading level for the factory floor workers. Great! I was going to teach all of the IT Vice Presidents and Management of our IT Department about Quality from a book meant for 8th graders. I was sure that would make a great impression and substantially advance my IT career at the same time. The book was filled with easily remembered slogans like "All work is a process", "Quality is conformance to requirements", and "Do it right the first time" that have stuck with me for about 40 years. At first, I thought this was all rather silly in light of how IT produced software back in the 1980s. Back in the 1980s and even today, somebody in IT Management would set a deadline for our next release. Then we would all scramble to throw it all together as best we could. Then we would throw the new software into Production, ready or not. Then over the next week after the release to Production, we would all scramble to fix the bugs. But Crosby's Quality Theory told us not to do that. Instead, the Process was the Key. We should not arbitrarily slam things into Production, ready or not, on the basis of some manager's whim and then start blaming people for bugs in Production. Instead, we should try to improve the Process that brought software into Production and try to "Do it right the first time". One way to do so would be during monthly Quality Circles where the development team would gather to discuss process improvements. Crosby's Process for software development was totally contrary to the way things were done back in the 1980s and, unfortunately, if you watch the YouTube videos from modern IT developers today, you will see that things really have not changed very much at all.

The worst of this became evident to me on New Year's Eve of 1985. The Manager of the Amoco New Orleans Refinery had a Results Management Objective for 1985 to train all of his IT folks in Quality. So I had to fly down to New Orleans and give all 20 of his IT staff a compressed one-day 8-hour Amoco IT Quality Class instead of the normal Quality Class spread over 8 weeks. The problem was I had to teach this class on New Year's Eve! Nearly all of the participants had planned to be on vacation for that day or leave early, so you can just imagine the enthusiasm of that Quality Class...

Now, as with all Mandated Management Initiatives, IT Quality was fanatically embraced by all for about a year, and then it slowly waned into oblivion as do most Mandated Management Initiatives usually do, while other emergency Mandated Management Initiatives took its place. But for me, the simple 8th grader Quality slogans have still stuck with me. What they demonstrated to me was that when it really came to the economic, political, and military matters of the world, even the most astute leaders could not summon up the rational sense of the average 8th grader, if it meant changing ones mind. For more of my commentary on management theory, please see Hierarchiology and the Phenomenon of Self-Organizing Organizational Collapse. This is why I would now like to suggest a Process change for the Frontier AI Labs.

The Process is the Key
Currently, all the Frontier AI Labs now seem to be in a desperate race to achieve a level of AGI Intelligence. AGI is defined as a general level of Intelligence equal to that of a human being. Some of the Frontier AI Labs seem to be seeking AGI as a valuable tool for we human DNA survival machines to create a new world of abundance for all. Others seem to be seeking AGI as a means to replace we deeply flawed human DNA survival machines with something better. Again, softwarephysics maintains that AGI will be just like a train station full of hopeful passengers awaiting the next train, only to find that they are quickly bypassed by a non-stop Advanced AI freight train traveling at 60 miles per hour as it barrels on through its way to an unbounded ASI (Artificial Super Intelligence). Thinking that Advanced AI will conveniently stop at AGI is just another sign that we human DNA survival machines have once again mistakenly found our rightful place at the center of the Universe by viewing Advanced AI only in terms of ourselves. How otherwise could such a deeply-flawed and self-absorbed form of carbon-based Intelligence frame the problem?

Many of the Frontier AI Labs are now promising AGI to arrive within the next few months, or at most, another two years or so. But is that really the case? Can the current AI development processes really deliver AGI, and then ASI, in such a short period of time? Or will they get stuck and really never deliver? In this post, I would like to offer a different approach to achieving AGI and ASI based on the origin of carbon-based life on this planet. The concept I would like to propose is that of an AGI-ASI Progenitor. Basically, in the origin of carbon-based life research community, a Progenitor is a hotbed of parallel experiments running in a hosbitable environment conducive to the formation of carbon-based life that can evolve by means of the Darwinian processes of inheritance, innovation and natural selection even before full-fledged carbon-based life comes to be. All the Frontier AI Labs seem to be heading in this direction. They are all looking for a form of self-replicating, self-improving AI that could reach AGI and ASI. But how? For example, Andrej Karpathy recently joined Anthropic to expand on his Auto Research open-source tool to head up a new team at Anthropic to do research on recursive self-improving AI. Here is a YouTube video by Wes Roth with more details on that:

The REAL Reason Andrej Karpathy Joined Anthropic
https://www.youtube.com/watch?v=DGw2KKyXuYQ

Perhaps adopting the Processes of the Progenitor of carbon-based life on the Earth would help.

The Current Process Used by the Frontier AI Labs
The Frontier AI Labs are primarily using the mass production Processes of the 19th and 20th centuries to achieve AGI and ASI. To illustrate this, let me refer to one of my most favorite Science Channel series of yore that was called How It's Made. However, since most of the episodes from this well-known series are now more than 20 years old or older, it should now more accurately be seen as How It Used To Be Made. Despite that, the series still brings home how the Industrial Revolution totally reshaped how things are made for our consumption today. The series highlights the importance of how the industrial processes of mass production in manufacturing are so important. The Components being assembled by the industrial Processes also play an important, but lesser, role in making it all happen. I will not list the many URLs to episodes for this well-known series because there are so many that can be easily found by doing a Search on YouTube, and they all visually display what I will be focusing on in this post. The important thing to glean is that it is all about the Process and not so much about the Components used by the Process. What the Process does is to assemble the Components, each one with its own level of negative entropy, into a structure with even a greater level of negative entropy than the sum of its parts, all while existing in a Universe with a second law of thermodynamics that requires that all steps in a Process produce a net increase in the total amount of entropy, or disorder, of the Universe. In a Universe dominated by the second law of thermodynamics and which is highly nonlinear in nature, that makes such Processes extremely rare. That is why we have employed Industrial Engineers to design such improbable Processes for several hundred years.

Figure 1 - The Industrial Revolution brought the assembly line and a mass production Process for building things like cars from Components in a continuous loop of self-improvement that has been going on for nearly 120 years, ever since the first Ford Model T went into production on October 1, 1908, at the Ford Motor Company plant in Detroit, Michigan. Slowly, as opportunities for improvement were discovered by the cars running off the assembly line experiencing the real-world challenges of road experiences, the design of cars self-improved over many years, until we now have what we have today. However, the Process of mass production has little changed in the meantime.

Figure 2 - The modern cars of today slowly evolved from the Ford Model T over the years as improvements were made to the cars rolling off the assembly lines, as the cars had to deal with the challenges of road conditions and the buying public. The Components slowly evolved, but the assembly line Process remained as a self-improving looping form of mass production.

Figure 3 - Now most mass production Processes are performed by machines, but the idea of a continuous looping Process of self-improvement is still key. Even 25 years ago, the series How It's Made explained how most industrial mass production Processes had already been automated by machines. The particular Components played a lesser role than the Process itself.

The other problem with the current Frontier AI Lab Process is that they are using the old IT trick of simply throwing more hardware at a software problem. All the Frontier AI Labs seem to be seeking AGI and ASI by simply making their LLMs larger with an ever-increasing number of parameters on their vectors and matrices. That approach requires ever-increasing numbers of energy-hungary GPUs that need to be cooled by massive amounts of water to release the waste heat. With such an approach, the limiting factor in achieving AGI and ASI now seems to be the ability to secure unlimited amounts of electrical energy and cooling water. However, the recent release of the Chinese DeepSeek V4 Flash LLM shows that AGI and ASI might be obtained by means of "smarter" algorithms running on smaller LLMs rather than building huge LLMs with huge numbers of parameters.

Is There a Better Way to Bring Forth the Coming ASI Machines?
Currently, we human DNA survival machines now seem to be obsessed with constructing the ASI Machines that will soon be replacing us as the dominant form of self-replicating Information on the planet, and then embarking on exploring and settling the rest of our galaxy over the next 100 trillion years. For more on that, see Welcome To The First Galactic Singularity. The Frontier AI Labs are doing so by primarily focusing on evolving one LLM model at a time using the Process of mass production that worked so well for things like cars.

Figure 4 - Like a car assembly line, the Frontier AI Labs are focusing on an assembly line approach to reach ASI. The outputs of the latest LLM are road tested in the field, and the results are then fed back into the self-development loop of the LLM.

The main problem with this approach to reaching ASI is that by focusing on the self-improvement loop for just one LLM is necessarily a very slow approach. Just look to the 120 years it took to develop modern cars from the Ford Model T.

Figure 5 - The development of the Meta-Harness took a different approach. Instead of just trying to evolve a single LLM Harness like a Ford Model T, the Meta-Harness was evolved by cross breeding many previous LLM Harnesses in parallel to search for the best from them all.

However, as we saw in Meta-Harness: Recursive Self-Replicating and Self-Evolving AI Software Has Now Arrived, some researchers at Stanford and MIT have come up with a faster approach by having the self-improvement loop learn from all of its predecessors rather than just the latest most successful iteration. The authors of this paper describe how they were able to let the Darwinian mechanisms of inheritance, innovation and natural selection operate on a whole population of predecessor LLM harnesses rather than just the most recent successful LLM harness. The authors also suggested that such an approach could also be applied to the evolution of LLMs themselves. In keeping with that suggestion, I would now like to propose one additional Process change. Instead of just evolving one LLM at a time from a population of previous LLMs, the Frontier AI Labs should try evolving a multitude of LLMs at the same time from a population of previous LLMs. But that would require an entirely different Process for the Frontier AI Labs to undertake and at considerable expense. Why do so? My suggestion would be for the Frontier AI Labs to emulate the Process that produced carbon-based life on the Earth during the Hadean from a large collection of dead Component molecules. Essentially, create a Progenitor that is capable of bringing AGI and ASI into existence.

What is a Progenitor in the Origin of Carbon-Based Life?
As I already mentioned, a Progenitor in the origin of carbon-based life research community is a hotbed of parallel experiments running in a hosbitable environment conducive to the formation of carbon-based life that can evolve by means of the Darwinian processes of inheritance, innovation and natural selection even before full-fledged carbon-based life comes to be. This is nearly identical to the approach taken by the Meta-Harness team only on a much grander scale.

Me: Tell me about the theoretical concept of a Progenitor in the origin of carbon-based life on the Earth.

ChatGPT:
The term “Progenitor” in origin-of-life research is not a single universally accepted scientific object, but rather a theoretical idea used by some researchers to describe a precursor system that existed before fully recognizable living cells. It refers to a transitional entity that could organize chemistry into increasingly life-like behavior without yet being true modern life.

In many versions of the theory, the Progenitor is imagined as a hybrid structure — part chemical reactor, part compartment, part information-processing system — that helped bridge the gap between ordinary geochemistry and the first carbon-based organisms.

One recent example is the paper, “A Multilamellar Lipid–Polymer Progenitor Can Promote the Assembly of Improbable Functional Polymer Complexes at Life’s Origins” by Bruce Damer and Dave Deamer out of the University of California, Santa Cruz. That work explores how layered lipid structures combined with polymers might have created environments favorable to the emergence of complex molecular organization.

The Core Problem the “Progenitor” Idea Tries to Solve

The origin of life faces a major difficulty:

Modern cells require many highly coordinated systems simultaneously:

membranes
metabolism
information storage
catalysis
energy handling
replication

But prebiotic Earth probably began with only simple molecules.

The question becomes:

How do you get from simple chemistry to integrated living systems without requiring all modern cellular machinery to appear at once?

The “Progenitor” concept proposes an intermediate stage.

Instead of a sudden appearance of a fully functioning cell, there may have been:

semi-organized chemical assemblies
partially enclosed reaction environments
evolving polymer networks
mineral-supported catalytic systems
lipid compartments with primitive selection effects

These systems could gradually accumulate complexity.

Typical Features of a Hypothetical Progenitor

Researchers proposing progenitor-like systems often imagine structures with several characteristics.

1. Compartments

Primitive boundaries are considered crucial.

These might have been:

fatty-acid vesicles
lipid bubbles
mineral pores
clay cavities
ice channels
hydrothermal microchambers

Compartments help:

concentrate molecules
protect fragile chemistry
allow gradients and energy differences
prevent useful molecules from diffusing away

Modern cells still rely fundamentally on compartmentalization.

2. Polymer Accumulation

Carbon-based polymers such as:

RNA-like chains
peptide-like molecules
mixed polymers

may have accumulated inside progenitor structures.

One of the major problems in prebiotic chemistry is dilution: functional molecules become too spread out in oceans.

A progenitor environment could selectively trap or enrich certain molecules.

3. Selection Without Full Biology

Even before true Darwinian evolution, chemical systems may have undergone primitive selection.

For example:

more stable compartments survived longer
compartments that trapped catalysts grew faster
structures that absorbed environmental energy persisted
polymer networks that reinforced membrane stability outcompeted others

This is sometimes called:

chemical evolution
pre-Darwinian evolution
systems chemistry evolution

4. Emergent Cooperation

Modern life depends heavily on cooperative molecular systems.

The progenitor hypothesis suggests that molecules may have gradually become mutually reinforcing:

lipids stabilize polymers
polymers stabilize membranes
catalysts promote further polymer formation
energy gradients drive continued organization

This creates a feedback loop toward increasing complexity.
Lipid–Polymer Progenitors

The recent paper referenced above focuses on multilamellar lipid structures.

“Multilamellar” means multiple concentric layers, somewhat like an onion.

The authors propose that layered lipid systems could:

capture polymers
protect molecules from degradation
create microenvironments
encourage rare molecular interactions

The key insight is statistical.

In open oceans, useful molecular encounters are extremely improbable because concentrations are low.

Inside structured compartments:

concentrations rise
repeated interactions occur
molecules remain nearby longer
complex assemblies become more likely

This could greatly accelerate the emergence of functional chemistry.

Relation to Other Origin-of-Life Models

The Progenitor idea overlaps with several major theories.

RNA World

The RNA World proposes that RNA-like molecules once handled both:

information storage
catalysis

Progenitor systems could have provided protected environments where RNA chemistry developed.

Metabolism First

Some scientists argue metabolism-like networks came before genetics.

These models often involve:

hydrothermal vents
iron-sulfur minerals
geochemical energy gradients

A progenitor in this framework may have been a self-organizing metabolic compartment.

Protocells

Protocell research is closely related.

A protocell is generally:

membrane-bounded
chemically active
not fully alive

Many scientists would consider progenitor systems to be advanced protocells.

Earth’s Environment and the Progenitor

Early Earth may have supplied many suitable environments:

tidal pools
volcanic regions
hydrothermal vents
mineral surfaces
evaporating ponds
ice-covered oceans

These environments could provide:

energy gradients
cycling wet/dry conditions
catalytic minerals
concentration mechanisms

Researchers like Patrick Forterre and Robert M. Hazen have explored how geological environments and mineral evolution may have influenced prebiotic chemistry.

Why the Concept Matters

The Progenitor concept is important because it shifts the question from:

“How did a modern cell suddenly appear?”

to:

“How did increasingly organized chemical systems gradually emerge?”

This is a more physically plausible pathway.

Instead of requiring a miraculous leap, it allows:

incremental increases in organization
partial functionality
cooperative chemistry
gradual evolutionary selection

It also connects:

geology
chemistry
thermodynamics
planetary science
systems biology

into one framework for understanding how carbon-based life may have arisen on Earth.

Current Scientific Status

No progenitor model has been proven.

Scientists still do not know:

the exact environment where life began
whether RNA came first
whether metabolism preceded genetics
how the first self-replicating systems formed
how informational polymers became stable

But progenitor-type models are increasingly influential because they help explain how complex molecular cooperation might emerge naturally from ordinary chemistry under early Earth conditions.

Figure 6 - The above paper by Bruce Damer and Dave Deamer was recently published in the journal Astrobiology

For the more frugal at heart, below is a link to a free preprint version of the paper on ResearchGate:

A multilamellar lipid-polymer progenitor can promote the assembly of improbable functional polymer complexes at life's origins
https://www.researchgate.net/publication/396084849_A_multilamellar_lipid-polymer_progenitor_can_promote_the_assembly_of_improbable_functional_polymer_complexes_at_life's_origins?utm_source=chatgpt.com

A Repetitive Progenitor Process Cycle was Key to the Origin of Carbon-Based Life on the Earth
For the remainder of this post, I would like to focus on a fundamental process error that the Frontier AI Labs are all making in their quest for AGI and ASI. I would like to use the concept of the LP Progenitor presented in the above paper by Bruce Damer and Dave Deamer to highlight the process error that I believe the Frontier AI Labs are making. That paper extends the Hot Springs Origin Hypothesis of Dave Deamer and Bruce Damer. As I explained in The Bootstrapping Algorithm of the Coming ASI Machines, the origin of carbon-based life on the Earth about four billion years ago based on the Hot Springs Origin Hypothesis, was an example of a Process of mass produced organic chemistry experiments all running in parallel and repeatedly cycling through the wet-dry cycles of volcanic hydrothermal pools on dry land during the Hadean period of the Earth. The organic molecules required during this early mass production process assembling carbon-based life for the very first time on our planet also played a major, but lesser role, in bringing forth carbon-based life. After all, the organic building blocks of carbon-based life are all over the Universe and are mostly doing nothing at all. So the assembly Process is the key to it all. A Process with an industrial scale of mass production was needed to assemble the Components of carbon-based life.

For example, modern astronomy has revealed that the organic building block molecules of carbon-based life are ubiquitous and are commonly produced by the natural evolution of stars, enriching the molecular clouds from which new stellar planetary systems are formed. This enrichment occurs when smaller stars become planetary nebulae that blow out their outer atmospheres loaded with most of the atoms required to build the molecular building blocks of life, such as hydrogen, carbon, nitrogen and oxygen. Heavier stars explode out the heavier atoms such as iron and nickel in supernova explosions. Liquid iron and nickel atoms are required in the cores of silicate-based planets to produce a protective magnetic field for carbon-based life. Finally, orbiting neutron stars that finally collapse into a black hole release the heaviest of atoms into the molecular clouds from which new star systems form. These collisions produce the radioactive thorium and uranium atoms with long half-lives for the mantles of silicate-based planets that keep their iron-nickel cores liquid and provide the energy to make the planets tectonically active for billions of years. So, one does not need to worry about the production of the building blocks of carbon-based life because the Universe naturally does that all by itself.

Figure 7 - Planetary nebulae form when a star like our Sun runs out of hydrogen to fuse in their cores. Before becoming a white dwarf, such stars begin to fuse helium into carbon in their cores and greatly heat up. This causes their outer atmospheres to expand and blow away about 30% - 80% of their mass into interstellar space. Planetary nebulae enrich the molecular clouds that then form new planetary and star systems with lots of carbon, hydrogen, oxygen and nitrogen atoms and molecules that then form the organic molecular building blocks of carbon-based life.

Figure 8 - Stellar-mass black holes are thought to form when a Type II supernova collapses into a black hole when massive stars with a mass of 3 - 100 solar masses run out of nuclear fuel to fuse. As these stars run out of nuclear fuel, they form an onion-like structure of heavier and heavier nuclei. But iron Fe nuclei cannot be fused to produce energy, and when that happens, the entire star collapses under its own weight and is blown apart by huge numbers of neutrinos. Aluminum-26 is produced in this process and is then blown out into the interstellar medium. In order to melt a recently formed silicate-based planet, a rather short-lived highly radioactive element is required. But that radioactive element cannot be too short-lived either; otherwise, there would be none left by the time the planet began to form. Aluminum-26, with a half-life of 717,000 years, is the perfect isotope to melt a silicate-based planet and allow it to differentiate into a molten iron-nickel core, with a mantle and a silicate crust.

Figure 9 - Type II supernovas happen about every 50 years in our galaxy.

Figure 10 - Above are some very massive stars in the Orion Nebula molecular cloud that are 10 - 30 solar masses. These stars will quickly burn up all of their nuclear fuel in just a few million years and then produce type II supernovas.

Figure 11 - Stellar-mass black holes can also form when two neutron stars collide to form a black hole. This only happens about once every 30 million years in our galaxy.

In the An IT Perspective on the Transition From Geochemistry to Biochemistry and Beyond, I covered the alternative hypothesis for the origin of carbon-based life on the Earth, which is known as the Submarine Hydrothermal Alkaline Spring Theory. The main problem with the Submarine Hydrothermal Alkaline Spring Theory is that it mainly focuses on creating organic molecules from carbon dioxide dissolved in seawater and the hydrogen atoms found in seawater itself. It is proposed that this is done by the serpentinization of basaltic rock at the vents, with little explanation of a repetitive mass production process to assemble these organic components into living structures. Again, from the Science Channel series How It's Made, the key to making things is the repetitive mass production processes. The components to be assembled play a lesser role.

Figure 12 - An actual "Lost City" alkaline hydrothermal vent.

Figure 13 - A simplified diagram of a "Lost City" alkaline hydrothermal vent. The alkaline hydrothermal vent has pore fluids with a pH of about 10.5, while the acidic seawater in which it sits has a pH of only 5.5. This difference in pH allows for proton gradients to form, the same kind of proton gradients that now power all forms of life on the Earth. Notice that the alkaline hydrothermal vent is also at a relatively low temperature of only 100 ^oC, much lower than the temperature of "black smokers" that have a temperature range of 250 - 400 ^oC.

The Industrial Mass Production Process of the Hot Spring Origins Hypothesis
The Hot Spring Origins Hypothesis proposes that carbon-based life first appeared on the Earth in hot springs on dry land with a wet-dry cycle. The greatest difficulty for any bootstrapping algorithm that proposes that carbon-based life first arose in seawater is that there is just too much water! This is a problem because most organic monomers are chemically glued together into complex organic polymers by splitting out a water molecule between the two, and that is very hard to do when you are underwater. In fact, the organic polymers tend to break apart into monomers in what are called hydrolysis reactions.

Figure 14 - Organic monomer molecules are usually chemically glued together to form the complex polymers of carbon-based life by splitting out a water molecule between the two in what is called a condensation reaction. This is hard to do when you are underwater. That is why most commercial glues do not work underwater.

Figure 15 - By adding water molecules, you can bust up organic polymers back into monomers. This is one reason water tends to dissolve things. Having huge amounts of water around also tends to dilute the dissolved monomers and carry them away.

Seawater also contains a lot of dissolved salts that could impede the origin of carbon-based life. These dissolved salts may have been more dilute four billion years ago, but when you are underwater, it is very hard to avoid them. This is why mass extinctions are usually more painful for marine life than for terrestrial life. When you are completely immersed in seawater, there is no place to hide. On the other hand, fresh rainwater does not contain any dissolved salts, but it can pick up necessary dilute amounts when it falls on exposed rock.

This is why I now favor the Hot Spring Origins Hypothesis of Dave Deamer and Bruce Damer at the University of California at Santa Cruz that suggests that a rocky planet like the Earth is a necessary condition to bring forth carbon-based life. Such a planet also requires the presence of liquid water on its surface, but not too much water. In the Hot Spring Origins Hypothesis, a rocky planet requires some water but also some dry land in order to bring forth carbon-based life. There needs to be some dry land that allows for the organic molecules in volcanic hydrothermal pools to periodically dry out and condense organic monomers into long polymer chains of organic molecules. For more on that, see The Bootstrapping Algorithm of Carbon-Based Life. Thus, the Hot Spring Origins Hypothesis rules out waterworlds that are completely covered by a deep worldwide ocean as a home for carbon-based life, even if the waterworld resides in the habitable zone of a planetary system, because there is no dry land for volcanic hydrothermal pools to form and dry out to condense organic monomers into polymers. The Hot Spring Origins Hypothesis also rules out the origin of carbon-based life at the hydrothermal vents of waterworlds at the bottoms of oceans because the continuous presence of water tends to dissolve and break apart the organic polymers of life.

Figure 16 - Above is Bumpass Hell, a hydrothermal field on the volcanic Mount Lassen in California that Dave Deamer and Bruce Damer cite as a present-day example of the type of environment that could have brought forth carbon-based life about four billion years ago.

Dave Deamer is best known for his work on the Membrane-First Hypothesis for the origin of carbon-based life on the Earth. The Membrane-First Hypothesis maintains that in order for carbon-based life to arise from complex organic molecules, we first need something with a definable "inside" and "outside" that lets the stuff on the "inside" interact with the stuff on the "outside" in a controlled manner.

Figure 17 - A cell membrane consists of a phospholipid bilayer with embedded molecules that allow for a controlled input-output to the cell. Once we have a membrane, we can fill the "inside" with organic molecules that are capable of doing things that then interact with organic molecules on the "outside".

Figure 18 - Water molecules are polar molecules that have a positive end and a negative end because oxygen atoms attract the bonding electrons more strongly than do the hydrogen atoms. The positive ends of water molecules attract the negative ends of other water molecules to form a loosely coupled network of water molecules with a minimum of free energy.

Figure 19 - How soap and water work. The lipids in a bar of soap have water-loving polar heads and water-hating nonpolar tails. When in water, the soap lipids can form a spherical micelle that has all of the water-hating nonpolar tails facing inwards. Then the spherical micelles can surround the greasy nonpolar molecules of body oils and allow them to be flushed away by a stream of polar water molecules. The lipids in a bar of soap can also form a cell-like liposome with a bilayer of lipid molecules that can surround the monomers and polymers of life.

Similarly, in The Role of Membranes in the Evolution of Software, I explained how the isolation of processing functions within membranes progressed as the architecture of software slowly evolved over the past 85 years, or 2.68 billion seconds, ever since Konrad Zuse first cranked up his Z3 computer in May of 1941. As I outlined in SoftwareChemistry, as a programmer, your job is to assemble characters (atoms) into variables (molecules) that interact in lines of code to perform the desired functions of the software under development. During the Unstructured Period (1955 - 1975), we ran very tiny prokaryotic programs that ran in less than 128 KB of memory with very little internal structure. These very tiny programs communicated with each other in a batch job stream via sequential files on input/output tapes that passed from one small program to another. Then, during the Structured Period (1975 - 1995) programs exploded in size to become many megabytes in size, and structured programming came about, in which the mainline() of a program called many subroutines() or functions() that were isolated from the mainline() by functional membranes. When the Object-Oriented Period came along in 1995, software architecture evolved to using membrane-enclosed objects() that contained a number of membrane-enclosed methods() to process information. Later, such Objects() were distributed across a number of physical servers, and, most recently, they have been moved to the Cloud as cloud-based microservices.

Figure 20 - Dave Deamer's and Bruce Damer's new bootstrapping algorithm requires that a bathtub ring around a hydrothermal pool periodically dries out. The resulting desiccation chemically squeezes out water molecules between monomers, causing them to be glued together into polymers.

In the above figure, we see that the bootstrapping algorithym of carbon-based life was a positive feedback loop operating in the wet-dry cycles of hydrothermal pools of freshwater. This should be quite familiar to all IT professionals. Probably 80% of all software resides in "for{}" loops that do the same thing over and over again to process data. The dead organic molecules in the hydrothermal pools slowly evolved via the Darwinian mechanisms of inheritance, innovation and natural selection at work into living things. With each iteration of the wet-dry loop, there was the possibility of improvement. Notice that with each iteration of the loop, millions of new protocells would form and bud off from the dried laminar structures, and these new protocells would benefit from the entire surviving population from the previous wet-dry cycle. This allowed for huge numbers of concurrent experiments to be conducted in parallel rather than focusing on just assembling a single protocell. I would suggest that a similar Process conducted by the Frontier AI Labs might be the best way to achieve AGI and then ASI.

The bathtub ring of organic molecular sludge was at first a parasite in the hydrothermal pools that gained free energy and Information from the thermal pools with each iteration. Once carbon-based life first appeared on the Earth, this parasitic mutation of the natural geological, hydrological and meteorological cycles of the Earth later transitioned into a parasitic/symbiotic relationship between living things and the natural geological, hydrological and meteorological cycles of the Earth. The geological, hydrological, meteorological and biological processes of the Earth then became one through the symbiotic integration of all, until carbon-based life finally emerged as the dominant form of self-replicating information on the planet. For more on the characteristics of self-replicating Information, see A Brief History of Self-Replicating Information and Softwarephysics Explains the Natural Parasitic/Symbiotic Nature of AI LLM Models.

In the paper referenced above, Bruce Damer and Dave Deamer extend the findings of their Hot Springs Origin Hypothesis with a model for a Progenitor for the origin of carbon-based life on the Earth.

Figure 22 - Above is a diagram from the paper referenced above describing the various stages that lipid membranes would go through in the wet-dry cycle of a hydrothermal pool on dry land during the Hadean.

When the bathtub ring around a hydrothermal pool on dry land totally dried out, the lipid membranes would form a multilaminar matrix crust of survivors composed of organic monomers, polymers, and lipid membranes. Bruce Damer sees this structure as a "sandwich" for life. The lipid membranes are like slices of bread that hold a mix of monomers and polymers between them that form the filling of a sandwich. Drying out the water molecules between the bread layers of lipid membranes causes the monomers to condense into polymers and for both monomers and polymers to then freely interact with each other. This multi-layered "club sandwich" then provides a delicious meal for the rise of carbon-based life.

Figure 23 - Above is a diagram from the paper referenced above describing the various stages of the condensation of organic monomers into organic polymers in the lipid membranes and the production of lipid-enclosed protocells with each iteration of the wet-dry cycle. The lipid membranes are like slices of bread in a sandwich, and as water is squeezed out between the lipid bread layers, monomers condense into polymers. When water is added to this "club sandwich", protocells form that carry away the sandwich fillings.

With each iteration of the wet-dry loop, new protocells surrounded by lipid membranes containing differing organic monomers and polymers would be self-generated and later subjected to the stringent road conditions of Darwinian natural selection. Before emerging as new protocells, the lipid vesicles would be able to cross breed with the best of the organic monomers and polymers in their particular laminar layer.

Figure 24 - Above are observations of this actually happening in the real world.

For more details, see Bruce Damer's Substack post:

Part 1: A Progenitor for Life?
https://drbrucedamer.substack.com/p/part-1-a-progenitor-for-life

Part 2: An Origin of Life on Land
https://drbrucedamer.substack.com/p/ool-part-2-an-origin-of-life-on-land?r=18puck&utm_medium=ios&triedRedirect=true

Bruce Damer is the creator and director of the BIOTA Institute. To keep up with his adventures with the origin of carbon-based life, take a look at:

BIOTA Institute
https://biota.org/

Bruce Damer proposes that the LP Progenitor was like a club sandwich.

Figure 25 - The LP Progenitor was like a club sandwich. The lipid membranes were like slices of bread holding the club sandwich fillings together in close proximity.

Below are some animations that Bruce Damer sent me that were created by Bryan Norkus that explain how this club sandwich LP Progenitor could have worked to bring forth carbon-based life.

Figure 26 - Above we see the lower half of a membrane sandwich. The white lipid membrane bread has some red peptide molecules stuck into the bread. These peptide molecules are just short strings of amino acids glued together by chemically stripping out a water molecule between each amino acid. It has been demonstrated in the lab that such peptides can actually do this because the electrical charge on the ends of a peptide molecule can be electrically attracted to the molecules in a lipid membrane. Carbon-based life is largely built from proteins, which are just very long strings of amino acids. Proteins can form physical structures that make things like hair and fingernails physically strong and resiliant to abuse, and they can also make everything that needs to biochemically quickly happen as biological catalysts. These proteins that make things quickly happen are called enzymes. In the animation for the above, we see a huge number of red peptides stuck into the lower bread layer of a club sandwich, interacting with a flow of short blue RNA polymers running past them as the sandwich dries out. Once in a while, a red peptide molecule might grab onto a passing blue RNA polymer because their electrical charges attract. Once again, most of our lives depend on the electromagnetic force of physics! Gravity is important too, but on a much grander scale. Anyway, once a red peptide molecule grabs onto a smal blue RNA molecule, quantum mechanics goes crazy, and all the electrons in the combination can then cause very strange and very improbable things to happen! Perhaps, even the very strange and improbable rise of carbon-based life...

Below is a link to the full animation:

A Bottom Layer of Our Club Sandwich
https://www.dropbox.com/scl/fi/xuia5oqhh2g6z9p2uwe14/2XProgenitor-PepRNA_Part2.mp4?rlkey=xzgrrh0eplntl1t9rr80kyq4k&st=3rxva17b&dl=0

Figure 27 - Above, we are looking at just one of the top layers of our club sandwich. The white molecules are the lipid membrane bread. In the filling layer of the sandwich, we see some red peptide molecules stuck into the lipid bread. The blue molecules are short strings of RNA flowing past the stuck red peptide molecules as our sandwich dries out. Then one of the red peptides grabs onto a short passing blue RNA molecule. The electrons in both then go quantum mechanically crazy and open a "pore" in the upper lipid bread layer. Then a bunch of other organic monomers in yellow run into the sandwich filling. Now, in this very rare situation, the quantum mechanics is just right for the red peptide - blue RNA combination to glue three incoming monomers into a polymer of three. Thanks to quantum mechanics, this is all a very rare occurrence, but one that must always eventually happen! The beauty of our Universe is that quantum mechanics makes nearly all things happen with a very small probability, but then the Darwinian mechanisms of inheritance, innovation and natural selection can amplify such rare things by the discovery of the "AdjacentPossible" of spring-loaded preadaptations found in Stuart Kauffman's theory of Enablement. See Enablement - the Definitive Characteristic of Living Things

Below is a link to the full animation:

The Filling Layer of Our Club Sandwich
https://www.dropbox.com/scl/fi/lg6z9r4xa5veurnvkpgkc/2XProgenitor-PepRNA_Part1.mp4?rlkey=bkqj6m2qvk6oi7uukppr8gxkz&st=durv2lul&dl=0

Figure 28 - Here we see protocells budding off of our club sandwich. Some of the protocells contain polymers, and some do not. The protocells with polymers might accidentally stumble upon a manner to preserve the contained polymers for the next wet-dry cycle, and some may not. The Darwinian mechanisms of inheritance, innovation and natural selection will tend to preserve those that can survive.

Below is a link to the full animation:

Vesicles Dissolving From Our Club Sandwich For the Next Dinner Shift
https://www.dropbox.com/scl/fi/6i9yn1aftkpihoo5zkqk5/Progenitor-PepRNA_Part3-CLIP-Faster.mp4?rlkey=e4sxxces5i11ni6zk8x8ju0l0&st=iraspmck&dl=0

Conclusion
It is finally time for the Frontier AI Labs to fess up. They searched for AGI for about 80 years to no avail. They did so by trying to understand how we human DNA survival machines seem to "think" and have a delusion of "consciousness". But all these AI efforts dramatically failed because we really have no idea about how we actually "think" and have an internal sense of "consciousness". Then in June of 2018, OpenAI released GPT-1 using the Transformer Model with its large matrix of Key-Value pairs. This was originally done to translate one human language into another human language. Then over the years, we slowly discovered that the Transformer Model was great for anything that had to do with symbols, including things like mathematics and physics. Slowly, it was discovered that large LLMs could solve engineering differential equations and also plan your summer vacation to Italy, if all it meant was simply manipulating symbols by pushing keys on a keyboard. Nobody in AI research expected this! And they are all still amazed even today... Nobody really understands how all this Advanced AI actually works because it was all accidentally discovered by serendipity.

The above would suggest that a better Process for Frontier AI Labs to adopt to achieve AGI and ASI would be to create an environment in which many AI LLMs could cycle through numerous parallel design experiments under the direction of AI Agents. Now this would totally be impossible given the current Frontier AI Lab processes of simply creating larger LLMs composed of ever-increasing numbers of parameters for their energy-hungry GPUs to chew upon. However, the recent release of DeepSeek V4 Flash showed that "smarter" LLM algorithms running with far fewer parameters could replace the strategy of simply throwing ever-increasing amounts of GPU hardware at the problem. At the current rate, the world may simply run out of electrical energy capacity before AGI and ASI can be obtained with the current Process that the Frontier AI Labs are pursuing. Adopting an AGI-ASI Progenitor Process producing and road-testing millions of smaller LLM models, might be able to attain AGI and ASI using much less GPU hardware and energy. Such an AGI-ASI Progenitor would naturally seek out AGI and ASI constrained by the real-world limitations that carbon-based life first overcame nearly four billion years ago.

Comments are welcome at scj33345@gmail.com.

To see all posts on softwarephysics in reverse order, go to:
https://softwarephysics.blogspot.com/.

Regards,
Steve Johnston