The Moore’s Law for everyone.

Accredited: Quan Nguyen Ha

“At the 2023 Forbes Business Forum (Vietnam) held a few days ago at the Gem Center, Mr. An Nguyen, the founder of Trusting Social (a leading AI company in Southeast Asia), discussed the concept of the ‘Agent Economy’ — a world where AGI Agents (Artificial General Intelligence entities) will perform a significant portion of human tasks. This process goes through five levels: (1) Communication AI (Conversational AI, chatbot solutions) (2) Deepening domain knowledge (Domain Knowledge — surpassing biases) (3) Job skills (Job Skills — executing skill-based tasks) (4) Self-learning (Self Learning — accumulating and synthesizing knowledge) (5) Full agency (Full Agency — capable of replacing humans in certain tasks).

These agents will transform all current consumers into ‘prosumers’ (professional consumers), each having AI doctors, psychotherapists, lawyers, and banker AI staff catering to their specialized needs. A scenario where knowledge work costs, intelligence costs, and energy costs all decrease significantly. Technology plays the role of a powerful equalizing force, making the world less unequal and more prosperous than its current state.”

The messages evoke associations with Sam Altman’s philosophy on AGI, the former chairman of the renowned Y Combinator startup accelerator. Founded by the legend Paul Graham, this organization contrasts and collaborates in attracting talent and young startup companies, similar to the Thiel Fellowship led by the Paypal Mafia group, headed by Peter Thiel. Sam co-founded OpenAI along with other notable figures in the tech industry, including Elon Musk (interestingly, the initial funding also came from Peter Thiel, with Elon later withdrawing from OpenAI). The Stanford network has fostered vital investment and technological collaboration connections. YC has produced over 3000 companies, among which are well-known names like Airbnb, Coinbase, Cruise, DoorDash, Dropbox, Instacart, Quora, PagerDuty, Reddit, Stripe, and Twitch.

Altman’s residence serves as the gathering place for a group of experts known as Covenant (named after the Hebrew scriptures — the promise that God made to Abraham). It’s a collection of leaders or specialists in significant fields like AI, robotics, cybernetics, quantum computing, synthetic biology, genomics, space travel, and philosophers. They frequently exchange insights about technological advancements, the nexus of human ethics, and the future scenario where AI and an enhanced version of humans compete or complement each other.

Sam, a survivalism/prepper advocate, left behind a concise 22-word statement warning about the AI threat: “Eliminate AI-induced existential risk, a global priority, along with the social risks of pandemics and nuclear war.” Powerful technologies can be perilously misused for humanity. However, no one can stem the flow of progress and the curious exploration of mankind. The convenience and advantages AI offers to humans are also immense (upsides). Sam’s vision, alongside the visions of Silicon Valley technocrats, about the future world, particularly in the United States, is succinctly depicted in the article below titled “Moore’s Law for Everything.” It delves into the socio-economic aspects AI can impact and suggests some proposals for humanity to confidently stride into the new era. Truly fascinating.

The work at OpenAI reminds me every day that significant socioeconomic changes could arrive sooner than we imagine. Software capable of reasoning and learning will increasingly participate in tasks currently performed by humans. Furthermore, power will shift from labor to capital. If public policies do not adapt accordingly, the majority of humanity will face worse outcomes than the current situation.

We need to design a system that embraces the technological future and fairly taxes valuable assets (the most valuable things in the world) such as companies and land, in order to ensure equitable distribution of the wealth about to be generated. This will lead to a less divided future society and enable everyone to compete fairly on the basis of these new technological gains.

In the next five years, computer programs with cognitive abilities will be reading legal documents and providing medical advice. In the next decade, they will be involved in assembly line tasks and may even develop emotional qualities like trustworthy companions. Several decades later, this system could undertake any task, including making scientific discoveries, thereby expanding our concept of everything.

No one can halt this technological evolution or the recursive loop of creativity. Smart machines, in fact, aid us in creating even smarter machines, thus accelerating the pace of evolution. There are three ensuing consequences:

1> This evolution will generate significant wealth. The prices of many forms of labor that constitute the costs of goods and services will plummet to zero as some powerful AI machine joins the workforce.

2> The world will change rapidly and resolutely, leading to a corresponding shift in policies for rational distribution of wealth and enabling more people to pursue their desired lives.

3> If we navigate the above two points correctly, the living standards of the majority will be better than before.

Standing at the starting point of a major transition, OpenAI has a rare opportunity to steer the course toward the future. These pioneering experiments cannot immediately solve present social and political issues, but they must be designed to bring about sensible social changes in the future. Policies that disregard or neglect these imminent transformations will fail, similar to how pre-agricultural or feudal societies collapsed in the past. The following guidelines will detail what is about to happen and the related plans to ensure our confident steps into the new horizon.

Part 1: The AI Revolution

When zoomed-out on the technology landscape over the scale of time, we see technological advancements following an exponential curve. Specifically, let’s compare the world 15 years ago (pre-smartphones) to 150 years ago (pre-combustion engine, pre-electrification), 1500 years ago (pre-industrial machinery), and 15000 years ago (pre-agriculture).

The upcoming changes align with some of humanity’s most remarkable abilities: thinking, creating, understanding, and reasoning. After the three technological revolutions related to agriculture, industry, and computation, we are now witnessing the fourth revolution — Artificial Intelligence (AI). This revolution holds promise for all if our present society manages this trend responsibly.

Technological advancements will make what we achieve in the next 100 years vastly greater than what humans have created before, reminiscent of the struggles to control fire or invent the wheel. We have constructed AI systems capable of learning and performing useful tasks. Though still in its infancy, this trajectory has started to become clear.

Part 2: Moore’s Law for Everything

Broadly speaking, there are two pathways to a good life: individuals earning more money (becoming richer) or prices decreasing (making everyone wealthier). Wealth can purchase power: how much we can acquire with the resources at hand.

The best way to increase societal wealth is by reducing the cost of goods, from food to video games. Technology will rapidly drive down the prices of many items. Consider the example of the semiconductor industry operating under Moore’s Law: over decades, chips (processors) consistently become twice as powerful while the cost remains the same every two years.

Over the past few decades, the costs of televisions, computers, and entertainment devices in the United States have significantly decreased. However, there have also been rising expenses, exemplified by housing, healthcare, and higher education. The process of wealth redistribution cannot be effective if these costs continue to rise.

AI will reduce the costs of goods and services by labor, which is a determining factor in costs at multiple levels within the supply chain. If robots can construct houses on your owned land and rely on locally harvested and processed natural resources like solar energy, for instance, the construction costs will decrease compared to hiring human labor. Moreover, if robots can be manufactured by other robots, the hiring costs will plummet much lower than when manufactured by humans.

Similarly, envision AI doctors diagnosing health issues more accurately than any human, and AI teachers analyzing and explaining what students couldn’t comprehend precisely.

“The Moore’s Law for Everything” should become a strong demand from the generation, particularly from those who face barriers (like income) to access certain things. It might seem like a utopian world in imagination, but technology can very well turn this vision into reality. Imagine a world where, after decades, everything from housing, education, food, to fashion becomes half the price every two years.

At that point, we will explore more novel jobs stemming from a technological revolution. With the abundance that the new scenario brings, we’ll have a bit more freedom to creatively decide what else to pursue.

Part 3: Capitalism for Everyone

A stable economic system demands two factors: growth and inclusivity. Economic growth is crucial since everyone desires to improve their lives year after year. In a zero-sum world, where there’s little or no growth, the democratic atmosphere can turn hostile as people seek to vote based on money rather than various factors. This adversarial state brings about distrust and strong polarization. In a growth-oriented world, aggressive confrontations are less prevalent, as individuals feel that victory can be achieved by all.

An inclusive economy ensures that everyone has reasonable opportunities to access necessary resources, enabling them to pursue their desired lives. This is highly important as it ensures fairness, nurtures societal stability, and generates large pieces of the pie shared for the majority while also evidently stimulating growth.

Capitalism is a powerful engine for economic growth as it rewards those who dare to invest in assets that generate substantial value over time, thus constituting an effective incentive system that fosters the creation and distribution of technological gains. However, the price of capitalism is inequality.

A bit of inequality is acceptable, and at times even crucial for driving societal development. Societies attempting to achieve perfect equality face various problems, such as losing the drive to compete and struggle. However, a society that doesn’t provide reasonable equal opportunities for all individuals would also struggle to endure in the long run.

The traditional method to address inequality is income taxation. For various reasons, this has proven to be ineffective and increasingly detrimental over time. Modern humans still need to perform jobs, but most of them don’t create economic value in the way we perceive value today. When AI engages in producing the majority of basic goods and services, humans will be free to dedicate more time to other things like socializing with friends, taking care of loved ones, appreciating art and nature, or engaging in social and charitable activities.

We should therefore focus on taxing capital instead of labor, and use this tax as an opportunity to directly distribute ownership and wealth to every citizen. In other words, the best way to improve capitalism is to allow everyone to benefit from it directly as true equity owners. This is not a new idea, but it becomes more attainable as AI becomes more powerful, generating wealth from it. There are two sources of wealth from this: (1) companies, especially those effectively utilizing AI, and (2) land, with a fixed supply.

There are various ways to implement this tax and concerns about the effective utilization of the collected resources. Over the long term, other forms of taxation will likely be phased out, and the vision described in the beginning of this article will become a reality (in the spirit of a conversation starter).

We can create what’s called an American Equity Fund. This fund would be capitalized from taxing companies at a specific valuation level, for example, 2.5% of their market value each year. Specifically, payable shares would be transferred directly to the fund, while the 2.5% tax on privately owned land value would be paid in dollars.

All citizens over 18 years old could receive an annual distribution, both in dollars and company shares, deposited directly into their accounts. People are empowered and encouraged to sensibly use these resources for anything they desire, such as education, healthcare, housing, or even starting their own companies. At this point, industries supported by the government with increasing costs would face strong pressure from customers (or users) who quickly seek better alternatives to serve their needs, due to a competitive environment.

As long as the nation operates well, every citizen will receive additional funds from the fund every year (of course, there will still be economic cycles of ups and downs). Thus, each citizen will have a larger slice of the pie: freedom, power, agency, and opportunities associated with economic autonomy. Poverty will be greatly reduced, and many individuals will achieve more of what they desire in their lives.

Tax payments in the form of shares are closely tied to the dynamics of both the company, investors, and citizens, in contrast to profit-based taxes. Creating these incentives is crucial as incentives are superpowers. Clearly, corporate profits can be disguised, delayed, or sent offshore, and they often disconnect from stock prices (anyone owning Amazon shares wants the stock price to rise). Personal assets become closely tied to the nation’s well-being, making their literal stake in the country a measure of how well the nation operates.

Henry George, a US economist and political philosopher, proposed the idea of taxing land value in the late 1800s. This idea gained widespread support among economists of the time. The value of land is elevated by what is built upon it: a network effect of companies operating on the land, coupled with public transportation infrastructure, enables people to access the land, along with nearby ecosystems of restaurants or cafes. The natural surroundings of the land also contribute to its increased value. Clearly, the landowner cannot perform all the tasks on it, so it’s fairer if its value is shared with the broader society.

If each person holds a slice of the pie from the process of American value creation, everyone desires this nation to improve: collective equity in the nation’s innovation and pursuit of success will be closely tied to (or nourish) individual motivations. A new social contract will lay the foundation for a more equitable society (in exchange for a ceiling for no one) and a belief that technology can create a virtuous circle of societal prosperity. We will continue to need strong leadership from the government to ensure the desire to drive stock prices remains balanced with environmental and human rights protection.

In a world where everyone benefits from capitalism as owners, collective focus needs to be directed towards creating a better society (more good) rather than less bad. The approach will have a larger impact than it may seem. Put simply, more good means optimizing the creation of the pie to be as large as possible, while less bad means distributing the pie as equitably as possible. Both approaches can increase living standards, but continuous growth can only happen as the pie expands.

Part 4 — Implementation Process and Issues

The amount of wealth available to contribute to the American Equity Fund is substantial, with its value estimated at around $50 trillion, based on the market capitalization of US companies. Assuming this is an average figure achieved over the past century, it is expected to at least double in the coming decade.

There’s also an estimated $30 trillion value of private land in the United States (excluding developments on the land). Assuming this value doubles in the coming decade — this growth would occur faster than historical rates — given the significant shifts the world has recognized due to AI, the value of land, as a limited asset, would increase at a faster pace.

Of course, if we increase the tax burden on landholding, its value will decrease proportionally to other investment assets. This is beneficial for society, as it makes essential resources more accessible and encourages investment rather than speculation. The value of companies may decline in the short term, but they will continue to perform well over time.

A reasonable hypothesis is that this tax would lead to a 15% drop in land and corporate asset values (and only take a few years to recover).

Under certain assumptions (current value, future growth, and value decrease due to the new tax), within a decade from now, every adult among the 250 million Americans will have around $13,500 annually. This dividend could be higher if AI accelerates growth, and even if not, the $13,500 would have higher purchasing power due to technology’s impact on decreasing the costs of goods and services. Additionally, purchasing power would significantly increase each year.

Companies will find it easier to pay the tax each year by issuing new shares representing 2.5% of their value. Clearly, this also motivates companies to avoid the American Equity Fund tax by offshore relocation, but a simple check related to the percentage of revenue originated from the United States can solve this issue. A larger concern with this idea is creating an incentive for companies to return value to shareholders rather than reinvest in growth.

Furthermore, if only publicly traded companies are taxed, it will incentivize companies to remain private. For companies with annual revenue exceeding $1 billion, we can tax them under an equity accrual model for a limited number of years until they go public. If they remain private for too long, we can have them settle the tax in cash.

Next, it’s necessary to design a system that prevents public voting based solely on monetary incentives, specifically an Amendment that strongly safeguards reasonable taxes that can follow this path. Importantly, the taxes should not be overly burdensome, as it would inhibit growth (specifically, the tax rate on companies should be lower than the average growth rate).

We also need a flexible system for quantifying the actual value of land. There are two methods: first, relying on the approach of powerful federal assessment agencies; second, relying on self-assessment by local governments, similar to how they decide property taxes. Naturally, the locally collected tax must be based on the assessed value in the same manner. Furthermore, if the market trading causes the asset value to change by a certain percentage, the government needs to reassess it upwards or downwards accordingly.

An optimized theoretical system would only tax the value of land, not the improvements on it. In practice, due to the difficulty in determining these values, we need to tax both the land and the improvements on it (at a lower rate to encourage combined value growth).

Lastly, we cannot allow the public to borrow, trade, or speculate on the future distribution of the fund, as it would hinder the time-based fairness distribution. The government must ensure that such transactions cannot be enforced.

Part 5 — Transitioning to the New System

A brighter future doesn’t seem overly complex: we need the technology to create more wealth and a rational policy to distribute it. The essentials must be kept affordable for everyone to access. As systems like the one above become popular, policymakers who embrace it sooner will gain popularity: they’ll increasingly resonate with the public.

During the Great Depression, Franklin Roosevelt implemented a massive social safety net that nobody thought possible five years prior. We are in a similar moment now. A movement that supports both business and people is likely to unite voters broadly.

To smoothly introduce the American Equity Fund politically (reducing transitional shock), the legislative process or policy setup needs to gradually move towards the 2.5% rate. The 2.5% rate only applies fully when the entire GDP increases by 50% at the time the law is ratified. Starting with small distributions will create motivation and comfort people with the new future. Achieving a 50% GDP growth might seem time-consuming (it took 13 years for the economy to grow to 50% like in 2019), but once AI emerges, growth will be super rapid. Complex land taxes will gradually decrease, focusing only on two fundamental assets as mentioned earlier (land and companies).

It’s almost impossible to stop the impending changes. Careful planning will help us create a fairer and happier prosperous society. The future is even better than we can imagine.