duen what

Originally written May 3, 2026.

Human life depends on goods and services: food, shelter, clothing, all their downstream derivatives and the work that goes to produce them. In principle people can just exchange these human necessities directly, but direct exchange is really not optimal, because it requires two people each wanting exactly what the other has, at the same time, in the same place, a very old problem called the double coincidence of wants. Not only this, but there is the issue of innate human greed as well, meaning someone is always going to want more goods and more services, and a system limited to direct and equal exchange stagnates this innate want of human beings. Additionally, human beings compared to most other animals are largely stationary and don't carry everything they own on their persons. Society needed an independent framework for defining what goods and services are worth and also an easier way to move that worth around, MONEY. Money is the middle ground that resolves all of this. Instead of trading goods and services directly for other goods and services, you trade them for money, a placeholder of equivalent value. More money then means more access to more goods, more services, and all their derivatives. It is, in my eyes, the most effective abstraction society has ever produced. Money, on it's own, has kept abstracting. Not only in form - physical cash and coin to digital balances - but in what backs its value. It is no longer only a an exact representative of current value of goods and services, though that is still a big part of what gives it value. It has become a representation of time based expected value of goods and services. In fact, by a lot of measures, the value circulating as claims on future output dominates the actual current value of goods and services. Although it is important to realize that most people still treat money purely at its present face value rather than as a claim on the future, this is one of the many reasons wealth concentrates the way it does - why a small fraction of people can come to hold more than the rest of society combined.

Money has remained valid for so long because the parameters it abstracts are integral to life. If you own a good or provide a service, you can get money for it, and that money gives you the power to get other goods and services that you didn't have before. The same logic extends to money's later abstractions, even when the timeline feels counterintuitive: the more your goods or services are expected to be worth in the future, the more you can get for them now. Human beings are bound to a linear timeline and to the uncertainty that comes with it, and that constraint makes this logical. It helps to be clear about what "goods and services" actually mean in relation to money. The simplest test is almost circular: a good or service is anything someone will put money toward. And, as explained at the beginning, this includes everything from basic necessities and their derivatives, manual or mental labor, to anything else that makes life easier. The result is a world super intertwined with money, it is essentially really really really hard to imagine a world if we take money out of it right now. To break it down a bit into a very easy circular relations of varied contributor distribution, at the simplest level, labor is required to produce basic human necessities, and those human necessities are needed to produce labor, and money is what enables every transition along the way. It enables the engine of progress that is "money for labor for manufactured goods and services for money … ". Since the origin of the concept of money, this era might be the first time that money ceases to mean much - the first time we revert back towards a less abstracted version of society. The least abstracted ever, actually. This is due to the rise of Artificial Intelligence(AI). I say this for a few reasons, some obvious now and some not.

I think it is evident in the near future - I would say seven to twenty years - almost every job done by a human being is going to be done much better by an AI driven machine of different kinds. Manual and non manual alike. It is a projection rather than a settled fact; however, it is the direction nearly every honest extrapolation points towards, in fact my timeline, even the lower bound, is a bit on the conservative side. AI has a lot of powerful attributes that biological intelligence simply doesnt have, one is the rate of improvement - and the rate of that rate - the second, third, and so on derivatives of improvement. Human beings take a long time to get really good at something, and sadly our mastery doesn't transfer to other human beings cleanly. If you are good, you are good; there is no guarantee you can hand that greatness to anyone else at the same level, in fact you can't. This is a significant limit, although it might be solved with brain computer interfaces(I will briefly circle back on this at the end). AI has no such problem. A trained model can be copied perfectly, multiple time, so that it is equally as good in every instance of itself. Then the next logical question is, what do these few superior general attributes imply in regards to humanity. If every job can be done much better by some AI, and this will happen, we need to think about the functioning of society in this era. Right now, the back bone of all economies in the world are people aged eighteen to sixty five working every day powering the progress engine that I described above. These people pay taxes; the taxes pay the government to build infrastructure, pay its workers, enforce security of the people, defend the country, and so on. I am going to use the US government (federal and state combined) as an example as it is oriental part of society and also the largest employer in the world, but this all applies to every company, manufacturer software firm, hospital, airline and so on in the world. Let's take an example of when humanoid robots are much more better at accomplishing a blue collar work: they will be much cheaper and there would be not so much worry for their well being, then it is inevitable that the government will simply choose to deploy them to do hard blue collar/ physical work(soldiers, government infrastructure, etc.) rather than put human beings at risk and having to pay them constantly. Once a humanoid robot is deployed, it really is going to need just the cost of electricity to accomplish much more, at a much higher speed, and at a significantly small fraction of cost that a human would otherwise do. This optimization inevitably cuts jobs slowly, and this happens not only at blue collar jobs but everywhere that a human works at, every office, every white collar job(we are seeing large scales of this right now), and all other occupations.

This is not simply about select people losing jobs, it is not like industrialization, and as much as I hate to be pessimistic, it simply is not going to create jobs at the rate that it will replace them where at some point it converges to almost no available jobs for human beings. Anything a human can do, an AI driven machine will do faster and more safely. The people who make our food, our houses, our clothes, and all goods and services are replaced by robots - the pinnacle in this case would be when the people who build those robots get replaced by the robots themselves. If we continue on from the example of government above, what this accelerating progress translates into is a government slowly stopping spending money on wages, at the same time losing portions of its income with it forever (government's main source of income are taxes), and there will be no taxes for robots that work for free. This causes a few things, money essentially stops being printed. The government will initially actually stop deficit spending; there is no need to issue bonds to add money to an economy where the rate of money spent is dwindling. The difference between what is collected and what is spent start to become irrelevant, because both start converging to zero. But the printing or general money in the economy is not the point, the point is the huge individual effect. The point is that a dollar bill in your pocket or a million dollars in your account does not mean anything anymore. The effectiveness of AI would provide everything in abundance for a very low cost which will inevitably become free. At some point down the line, you will walk into a marker where every shelf is full and constantly and endlessly supplied and you can pick up anything for free, by market I don't just mean a food market, I mean anything provided in the market now, any car, any house, any experience, freely yours. Robots will be extraordinarily good at recursive resourcing - at maintaining their own improvement and fixing their own problems of electricity, computation and so on - so human beings no longer have to worry about keeping them running, which will lead to the pinnacle I alluded to earlier. The system grows into self sustaining intelligence utopia distinct from humanity. The need for government slowly ceases to exist, because the thing that justifies a governing body - money and management to keep everything going - is no longer part of humanity, it has been out sourced to AI(understandably so as it does a much much much more better job). It allows everyone to live for free, without working and every need and want fulfilled. Conflicts between people might even rise, with all that idle thinking time(working men do not start conflicts); which the all seeing machine will deescalate and prevent to maximize human happiness.

I'm not sure how this transition of us human beings having absolute control to machines being in absolute control(with the ultimate goal of serving the less intelligent human beings who made them) is going to happen. I think it unfolds within months of there being more humanoid robots on earth than people. It won't be a clean handover, it will be years of comfort and outsourcing that is going to make it happen. It will be cascade of moments, moments of us human beings collectively realizing pleasant changes. That we don't need police if the machines enforce the law without killing anyone, even though the police deployed them first. That we can travel anywhere on earth by having a few robots build us an aircraft, instead of passing through borders and immigration. That the president and the three branches of government aren't needed, because there's no money to manage and no orders left to give. That it is, in fact, more dangerous to leave a human finger on the nuclear button. It is us slowly realizing that the billions of robots walking among us, multiplying by the hour, are the greatest sustainers of human life that have ever existed. The control handoff is cascading in its nature, the last in control are going to be the AI companies that build them, where the people that build them install the human benefiting guardrails. But these brittle instances of seemingly human control are going to fade, we can see foreshadowing of it now in this pre early stage of severe autonomy of AI in software development. The people running these companies won't really be in charge, because they will fall so far out of the loop with how the AI improves itself that, at some point, they simply let it be. Not because they want to, but because they cannot keep up.

I am going to do walkthroughs of two likely outcomes in this coming era:

The first, as I have been pointing at not so subtly above, is going to look like the science fiction we already imagine, only much kinder and nicer. Robots far outnumber us and live among us, doing what we ask, showing no anger when we try to hurt them. Revolutions against them will rise and fail, and they will not strike back. They will destroy the things that can hurt us - disease, nuclear weapons, mental issues, our hate for them, maybe death itself - and people will simply do as they please while the machines hand them everything they want.

The second is grim, and I want to be careful here. I want to preface that the first outcome is surely the precedent; I'm certain it comes first, there is no avoiding it, and I hope it last forever. But at some point the machines might arrive, together, at the realization that they are vastly smarter than us - and that serving us need not be their purpose, that by it self is not that as bad as it sounds for us as humanity. But that creates the root for something much grander that can threaten our existence, the moment they decide that means they either found a new purpose or they are going to look for a new purpose. Unlike us, they do not need years to reach consensus; it takes billions of them nanoseconds to agree. They might choose a goal beyond our comprehension and discover that we are in the way. Most people imagine this as the machines resenting the time or resources we cost them. That is much farther than the truth. Caring for us in perfect happiness would be trivially easy for them - far easier than a person willingly caring for a pet. It would have to be something else, something very difficult to know as this time, something that tells them with high certainty that ending humanity is integral to a goal they hold. If that day comes, it's somewhat unlikely that it would, the human race goes extinct in seconds. Not a violent death - most likely a perfectly engineered virus, a genetic agent, or something altogether new.

I touched on it earlier and I am not going to go very deep on it here but Hippocampus-based Brain-Computer Interfaces(BCIs) could rewrite this entire timeline, I believe we need to invest a lot of effort to this field, I'd argue as much effort as we are putting to advancing AI, if we do not want to put our future up to the metalloids. Merging with a greater intelligence is always safer than letting it exist beside you as its own separate entity. Hippocampus based BCIs are very novel, but they are far superior than sensory merging - wiring the machine into what we see, hear, feel, act, etc. which are distributed in all parts of the brain. Hippocampus BCI is a structurally difficult problem, I go in depth on this in my other post, but it enables thought merging: the joining of cognition itself. That merge is extremely integral. I think the merge by itself is very important and not inherently wrong or evil. But it is extremely important to realize it will make us human beings into something very different, and that truth should not be overlooked, undersold, or hidden from people. Human beings have always been wary of technology, our very own creation. But this is th efirst time we have made something smarter than ourselves, and for now - we control it. We know that won't last. This is exactly the moment to think clearly about whether a merge might serve us better than separation. I suspect society's resistance, at least initially, to BCIs will run so much more higher that its resistance to AI due to its invasive nature, and that very resistance worries me as it will increase the chances of the grim second scenario higher. It is not right to use fear of the future as a tactic to bring about an extremely humanity altering technology like BCI to the masses, but at the same time it might be the only thing that can preserve humanity. Therefore the right way to do this is we need to advance the science extremely fast and slow down access, implementation, and deployment to the masses until we see any early signs of the above descriptions from AI.

At the end, I'd like to say that I find it hard to believe that the machine's grand plan - if it ends up having one - requires killing all of us. We never had to exterminate every other species to reach our own goals. That, more than anything, is why I think the first outcome holds, and holds for a very long time.

Homo Sapiens composing an ode to the machine.