The Hacker Paradigm: Open Source Innovation in a World of Plenty

Abstract

This paper explores the future of artificial intelligence and automation. After considering various views on whether AI and automation are beneficial, the paper argues that if work as a social institution becomes fundamentally obsolete, a new hacker paradigm is needed to provide "qualitatively rich opportunities" for people to exercise their agency. That is, because work plays a fundamental role in human development, other opportunities are needed to fill the gap. The paper argues that open source ecosystems are essential to achieving this goal. The paper goes on to argue that potential AI monopolies like Google and Amazon will shrink the space for people to find meaningful work in a rapidly changing economy.

Special thanks to Eric Zhang, founder of DoraHacks, for insightful discussions and support.

Part 1

The wave of automation is undoubtedly coming. Human labor will rapidly devalue. Artificial intelligence has already made great progress, and, as many innovations reduce costs, more and more companies will be able to integrate this technology. As production methods, capital, and supply chains are increasingly automated, productivity and efficiency will accelerate.

The question is whether this is a good thing. The answers vary. From doomsayers such as Eliezer Yudkowsky [1] and Roman Yampolskiy [2], to those who believe that artificial intelligence has the potential to - and will eventually - wipe out the human species. Humans will be far less efficient than future artificial intelligence, so when general artificial intelligence (AGI) is achieved, intelligent machines will easily conclude that humans consume too much and the world would be better off without them. Then there are the skeptics, such as Max Tegmark [3], who argue that there should be a moratorium on the development of AI in order to better align it with human interests. “Powerful AI systems should be developed only when we are confident that their impacts will be positive and their risks manageable.” [4] Then there are the technological optimists, such as Mark Andreessen [5], Sam Altman [6], and Ray Kurzweil [7], who argue that AI will usher in a material utopia. Development should be accelerated, regulators should back off, and the future will bring economic growth, increased productivity, and astonishing improvements in living standards.

It’s hard to make specific predictions here. People on different sides often tout their views with a certain tone of “inevitability.” Andreessen is one of the most prominent advocates of optimism, convinced that the future brought about by rapid technological progress will be the kind of future that every generation has dreamed of. There will be AI tutors with infinite compassion and patience, making education more universal and accessible; the medical industry will be greatly improved, providing better predictions and reducing errors; robots can perform dangerous jobs, allowing people to pursue their desires. The economic benefits will also be huge. Due to the reduction in production costs, prices will be lower, people's money will have more purchasing power, demand will increase, new jobs will be created, and wages will increase. His main argument for this view is historical precedent. Every time a new technological revolution comes, society always gets a net good from it. Before the formation of the oil industry in the 19th century, the whale industry employed thousands of workers. Oil replaced these jobs, but also eliminated the need to rely on slaughtering thousands of whales to make everyday products. The car threatened many industries, but when it became a commonly owned property, suddenly more roads, bridges, and gas stations were needed to meet the demand. Although there will be automation, AI will flood the world with good results, giving the economy unprecedented opportunities for people to find jobs, accumulate wealth, and take advantage of the value created by AI.

However, AI is different from previous technological developments. Artificial intelligence is the advancement and spread of intelligence, not just mechanization. Kurzweil writes: "The first industrial revolution expanded the capabilities of our bodies, the second expanded the capabilities of our minds." But a new variable means a different outcome. AI can learn, adapt, generate, discover, refute, doubt, confirm, set goals, formulate means, and perhaps one day have will and feel emotions. It is becoming increasingly agentic and autonomous, able to take on and complete tasks generated by its own processes. If it is not agentic now, the goal is to get it there. Predictions based solely on prior experience can be a category error, prone to misleading and error. Of course, there is overlap between past forms of technology and artificial intelligence, but there is a clear distinction between them, and that distinction should guide our predictions.

On the other hand, the views of eschatologists do not have much practical significance. The "Terminator"-like scenarios that often lurk behind eschatologists' warnings are unlikely to happen. Although, as Elon Musk[8] recently suggested on The Joe Rogan Podcast, if the wrong goals are embedded, perhaps a scenario like HAL 9000 in the Space Odyssey series[9] is possible. Given DeepSeek’s recent progress in reducing computing costs to the point where its AI models are now competitive with the most advanced models in the United States, a new race has begun and things are going to accelerate. The US Vice President’s speech at the AI ​​Action Summit made this clear. There is a race for international technological dominance that will shape the future, and slowing down development is tantamount to raising the white flag. Therefore, if there is any viable objection to Anderson’s claim, it must focus on how to guide technological development, not stop it. This should provide implications for the discussion about automation: how can automation be best guided to avoid undesirable outcomes? We need a proactive approach.

Section 1.1

One ​​problem with Anderson’s optimism about automation is that he is too certain that new jobs will be distributed enough for people to enjoy and take advantage of them. The problem is that the economy is becoming increasingly reliant on knowledge and information. Knowledge and information far outweigh manual labor, and the people who succeed in the market are those who have the relevant skills, which are acquired through education and training. As Thomas Piketty [10] observes in his book Capital in the Twenty-First Century, this is a general economic trend: “The main driver of convergence (i.e., the reduction and limitation of inequality) is the spread of knowledge and investment in training and skills.” Thus, as the economy becomes increasingly reliant on knowledge, unemployed workers without the relevant skills or security will be severely disadvantaged. Those who have the relevant knowledge, skills, and information will be better placed to invest capital and increase their wealth. For example, even with economic growth and lower production costs, those without skills will see wages stagnate. But assuming that demand could rise due to lower production costs and greater purchasing power for consumers, prices could rise if supply cannot meet this demand quickly, triggering demand-pull inflation. The key is that there are variables and scenarios that economic and productivity growth may not naturally cope with. As Piketty says, “There is no natural, spontaneous process that prevents destabilizing, unequal forces from permanently taking hold.” Therefore, proactive response strategies must be adopted to avoid the sharp rocks that could sink the many ships that float on the surface of the economy. In his book The End of Work, Jeremy Rifkin [11] takes a more somber tone: “The [next] industrial revolution is forcing the global economy into an unprecedented crisis, with millions of people losing their jobs to technological innovation and global purchasing power falling sharply.” In The Second Machine Age [14], Erik Brynjolfsson [12] and Andre McAfee [13] make good reasons to think that Rifkin’s view has some merit. They argue that the bounty of economic growth—that is, producing more output from less input, making goods cheaper and more available, and thus improving people’s lives—is not benefiting everyone by distributing wealth, income, and capital in a way that benefits everyone. In other words, “the tide of technological progress may not lift all boats.” For example, on the skills gap mentioned above, “since the mid-1970s, wages for people with graduate degrees have risen by about 25 percent, while the average wage for high school dropouts has fallen by 30 percent.” Despite the “impressive trajectory” of U.S. GDP and economic productivity since the mid-20th century, the country’s median income has fallen, suggesting that a few are reaping the benefits while the majority are missing out: the gap between the rich and the poor has widened since the Reagan years. This Brookings Institution article [15] supports this view.

However, perhaps the practical implications of these views are misguided by misplaced sympathy. Although inequality does exist, that does not mean it should be considered the primary variable of concern. Again, technological innovation has made goods incredibly cheap, made people’s lives easier, raised living standards, and made daily life more convenient by providing greater access to goods, services, information, and knowledge to meet people’s needs and desires. Although wealth may be unequally distributed, its overall growth has made everyone better off. And, as wealth grows further, this trend will continue. Things that were once only affordable to the rich are now affordable to most people. For example, the iPhone has more information than all the existing libraries, and historically, only the richest have access to the latter, while almost anyone can buy the former. The largest library in Europe was the Vatican Library in 1481, which contained an estimated 3,500 books and documents. ChatGPT estimates that if the information on the Internet were converted into book-sized quantities, it would be about 467 quadrillion books. Look at the development of agriculture. When the economist Milton Friedman wrote Free to Choose in 1980, he noted that in the era when the Declaration of Independence was signed, “19 out of every 20 workers were needed [to work in agriculture] to feed the country’s inhabitants and to produce a surplus to export in exchange for foreign goods. Today, less than 1 out of every 20 workers is needed to feed the country’s 220 million inhabitants and to produce a surplus to make the United States the world’s largest exporter of food.” Peter Theil [16] often claims that the problem is stagnant technological innovation, not moral failures related to welfare programs or distribution models. Science and technology need breakthroughs to transform industries and raise living standards. The intelligent revolution of artificial intelligence should be able to realize this potential. So perhaps it is best to just leave it alone. Focusing on fairness or justice in a particular pattern of distribution is, as Nietzsche put it, a “bad-conscience”—the guilt experienced when pursuing personal gain at the expense of the perceived collective good, but in reality, only serves to hinder progress rather than promote it. The true and best goal, arguably, is to strive to innovate and create the values ​​that drive history, which naturally tempers people’s self-interest, as Adam Smith’s [17] metaphor of the “invisible hand” suggests. As Milton Friedman put it, “No external force, no infringement of liberty, is necessary for the state of things in which individuals cooperate, so that each of them may benefit from the cooperation. That is why, as Adam Smith put it, the individual ‘who only cares for his own interest’ is ‘guided by an invisible hand to promote an end foreign to his own purpose.’”

Section 1.2

The question of whether economic growth is sufficient to improve long-term well-being and living standards is highly controversial. This question is obviously not addressed here. However, the point of mentioning this is to show that both views have validity. Individuals pursuing their own interests in the market, with variables like prices and wages balanced through voluntary exchange among people, have brought great results. But gaps have also emerged. Automation is a potential threat. It is a mistake to ignore it and claim that it is alarmist. The transition period after possible AI disruption and before some form of balance is achieved may be complicated and may cause unnecessary pain that could have been avoided if it was dealt with in advance. Revolutionary new technologies may have both good and bad sides. For example, the printing press in the fifteenth century freed people from the control of information imposed by the church, giving rise to a wave of scientific literature that provided great value to civilization. But it also led to fierce and bloody religious wars because it allowed people to spread all kinds of information. Therefore, in order to avoid similar results, it is important to take a proactive approach when implementing any technology that can rearrange and reorganize socioeconomic arrangements. From this, the usual inference is that regulatory measures are the best means to deal with these possibilities. But rather than settle for regulation, which, as Marc Andreessen warns under a Biden administration, would lead to AI monopolies, this article will seek to support voluntary exchange in free markets while promoting the social cooperation necessary to help people adapt and function in an economy that is increasingly less dependent on human capital.

The concern here is that people will lose the opportunity to exercise their agency, and thus have a sense of control and influence over their lives. If large numbers of people lose the opportunity to participate in work, a gap will appear that blocks the way people usually realize their potential in the world. They will no longer have the opportunity to make life choices with far-reaching consequences in key social areas, choices that require care and foresight. People will lose the opportunity to test themselves, discover their natural inclinations and interests, identify what they value, and exercise choice to pursue it. The main opportunity in society to cultivate perseverance, motivation, and personal will will be lost. If work is to be largely replaced by automation, then a new medium of opportunity will need to take its place. Work as a social system requires progress and maintains avenues for people to grow and develop.

John Dewey, a 20th century liberal philosopher, wrote: "Just as the senses require perceptible objects to stimulate them, so our powers of observation, recollection, and imagination do not work spontaneously, but are stimulated by the needs set by current social activities." Social environments like education and work provide the medium of activity needed to stimulate specific abilities and functions required to complete the tasks that define the activity. It is often argued that automation will free people from the constraints of tedious and mechanical work. They will have more time to find new ways to express their needs and eventually achieve the goals that work has been preventing them from achieving. People will have the freedom to realize their potential. But this will not happen spontaneously. There needs to be an environment that stimulates activities that achieve true freedom, meaning, and purpose. When a person is completely alone in a room where no one hinders his movements and inclinations, he can be said to be free. Such a person can do whatever he wants. But without situations that can provoke appropriate responses, without the possibility of success and failure, without scenarios that require cooperation and coordination, which help to develop social and emotional abilities, and without the need to exercise intellectual and problem-solving abilities, the growth of the people in this room will stop, and they will wither and fail to achieve freedom. A person who falls freely from the sky, unimpeded by anything, may seem free, but he is not free, because he is destined to hit the ground. As Franz Kafka wrote: "I am free, therefore I am lost." The philosopher Jean-Paul Sartre said: "Human beings are destined to be free." The question at hand, therefore, is how to open up a space into which people can voluntarily enter and find opportunities to express the abilities needed for a good life. Such opportunities, aimed at exercising people's agency, will be called "qualitatively rich opportunities", and the environment that enables these opportunities will be called "qualitatively rich environment".

A “qualitatively rich environment” that embodies “qualitatively rich opportunities” is *“open source ecosystems”*, which should be the primary model for AI development, providing the following framework: The market will provide opportunities for people to exercise their agency through challenging and stimulating work that leads to meaningful projects. People will have to learn new skills and overcome obstacles, but the process is not only to maintain a market where people can compete, make choices, and develop agency, but also to promote purpose and meaning in people’s lives. If the economy is to close the door on old forms of work, such as assembly line and service work, it should provide more opportunities for people to do meaningful work. As economist Tyler Cowen pointed out in the Lex Fridman podcast, people need things to do. They don’t want to just stay at home, as the COVID lockdowns have demonstrated. If routine and low-skilled labor is to be automated, the only option is to ensure that there are “qualitatively rich opportunities” so that people can use and exercise their abilities to perform meaningful work.

Therefore, a new paradigm should be introduced in the economy. Drawing inspiration from Pekka Himanen’s [18] book The Hacker Ethic, people should move from an industrial-based work paradigm to a “hacker paradigm.” In the industrial paradigm, individuals mainly perform repetitive labor tasks with the sole purpose of earning income to survive. The hacker paradigm, on the other hand, is centered on intrinsic value and emphasizes the fun, excitement, creativity and playfulness of work itself. It is characterized by passion and play rather than the social obligation to “fulfill one’s duty to work” or “contribute to the economy,” as Max Weber [18] revealed in Capitalism and The Protestant Ethic. The hacker paradigm emphasizes human agency and its potential to realize values ​​that elevate humanity. The industrial paradigm creates a sense of alienation between workers and their active participation in the world: workers are indifferent to the work they contribute; people often dislike or even hate their work. They are there to make money rather than to participate in meaningful work. As Shimanin writes, “Reforming the form of work is not only a question of respecting workers, but also a question of respecting humans as humans. Hackers do not subscribe to the maxim ‘time is money’ but to the maxim ‘it’s my life.’” As Eric Steven Raymond puts it in How To Become A Hacker [20], hackers see the world as full of wonderful problems and find freedom in working to solve them. They seek out projects that require motivation and passion, and believe that the world will be infinitely better when others have the opportunity to have the same freedom, which requires social cooperation and the greatest possible open access to information—which is the significance of “open source.”

To highlight why the introduction of a new paradigm is essential to meet the challenges of artificial intelligence and has significant historical significance, we can return to the ideas of Adam Smith. Later, the concepts of "qualitatively rich environment" and "qualitatively rich opportunity" will be further clarified and applied to the current system of artificial intelligence monopolies centered on proprietary technology. It will then be seen that these monopolies will stifle the generation of the hacker paradigm and are therefore undesirable.

Through the perspective of Adam Smith, we can see that although industrialization has greatly improved people's living standards in the long run, it has also had a negative impact on people engaged in repetitive and mechanical labor, because such labor weakens the stimulation of human cognitive abilities and hinders their full development. The division of labor has greatly accelerated the process of specialization, but it has reduced individuals' exposure to opportunities that can mobilize their higher mental abilities - these abilities are the basis for achieving meaningful, purposeful, and fulfilling work. Artificial intelligence and automation are becoming a catalyst for transformation, gradually replacing the work models and production logic of the industrial era, and transferring tasks that human capital once performed to intelligent machines. If artificial intelligence is truly an advancement - that is, human advancement - then it should respond to Adam Smith's concerns about the division of labor and its impact on human life.

Part 2

Section 2.1

Adam Smith wrote The Wealth of Nations in 1776. Influenced by Isaac Newton, Smith wanted to understand the design behind society and the mechanisms by which it worked, similar to how Newton discovered and used mechanistic terms to formulate the physical laws that explained the universe. The model of that era was to see phenomena as organized and ordered like a finely tuned watch, with all the parts and mechanisms working together in perfect harmony to produce a variety of effects and emergent phenomena.

During Smith's lifetime, the first industrial revolution was underway in England. Cities expanded, goods and services once reserved for the wealthy became cheaper and more accessible, living standards improved, infrastructure developed, and on the surface, people lived better lives than ever before. Smith observed social progress. As a result of the Age of Discovery and the technologies that enabled Europeans to set foot on new continents, Europeans came into contact with peoples that seemed to have not yet undergone the historical processes that Europe had experienced and were considered "uncivilized" by Europeans. Therefore, the ethos of the continent was that history and civilization were entering a new and more advanced stage. Innovations brought about by scientific exploration and knowledge increased productivity and efficiency, thereby making social systems more complex to organize. Smith wanted to understand the causal mechanisms that drove this process.

Smith observed and believed that the division of labor played a major, even fundamental role in social economy. As social organization expanded, manufacturing and production processes were gradually broken down into more specialized tasks, and as a result, the output of these processes increased significantly. In the traditional guild framework, craftsmen devoted their lives to a complete craft, learning all the aspects required and mastering each part. This guild framework was gradually replaced by the specialized factory model, which systematized production through the division of labor in the growing city. A large system of individuals emerged, which required a great deal of coordination, and the division of labor organized this system into a complex mechanism designed to reconcile people's interests. Smith also believed that the division of labor reflected some deeper characteristics of human nature. People flocked to cities to earn wages for their labor and enjoy better living conditions because, Smith believed, humans were naturally inclined to exchange (truck), barter (barter), and trade (exchange). [21] "The disposition to exchange, barter, and to trade one kind of thing for another," he wrote, "is common to all mankind, and is not to be found in any other group of animals." Animals like wolves naturally organize themselves in hierarchical groups, newly hatched birds tend to fly, and organisms like trees compete for sunlight in dense forests. In contrast, humans exchange, barter, and trade because they have an innate tendency to improve their conditions. The division of labor facilitates this goal.

However, Smith believed that the division of labor was a trade-off—it had a cost. Although people's living standards improved and they enjoyed better conditions and were able to meet their consumer needs such as food, water, and shelter, the division of labor reduced workers to cogs in a machine. These tasks required little cognitive effort. Once the tasks became routinized and transformed into muscle memory, there were no more obstacles or problems to overcome. As a result, Smith was deeply concerned that this process would have profound negative cognitive effects over time. People went to work, and as long as they were familiar with their tasks, they did not need to use their brains; they acted mechanically. Smith wrote bluntly: “A man who has been performing all his life a small number of simple operations, whose effects are always evident, has no need of exerting his understanding, or of exercising his ingenuity to seek solutions to difficulties which never arise. He naturally becomes almost as stupid and ignorant as it is possible for mankind to be.” The opportunities provided by the division of labor are qualitatively poor, and fail to equip people with the abilities to lead meaningful and purposeful lives. To borrow a term from another Enlightenment philosopher, Immanuel Kant, the division of labor fails to respect human dignity, which requires targeting and cultivating the abilities necessary to enable people to lead lives of their own choosing. Having dignity means having influence and control over one’s own life. The division of labor creates more opportunities to improve living standards, but it does not provide the “qualitatively rich opportunities” that allow people to integrate their abilities and achieve a well-rounded life. The opportunities available to most people do not encourage them to focus their creative and productive capacities on goals that would expand their sense of freedom in the world. The division of labor offers more opportunities than the guild system, but these opportunities are also shallower and less likely to generate experiences of mastery. Yet because living standards are higher, people rightly choose jobs that relegate them to cogs. But as Smith observed, the rise in living standards is insufficient to meet the full range of human needs. It does not enable most individuals to exercise their higher capacities.

Improvements must be made, Smith argued. It is essential for people to have more “qualitatively rich opportunities” that enable them to live lives of greater meaning and purpose, that embody human dignity, and that provide a basis for the exercise of their agency. The division of labor offers gains in material freedom, but it does not offer enough opportunities for a more intrinsic, agent-driven freedom.

This is surely one of Karl Marx’s great critiques of capitalism. Under capitalism, people are alienated from their labor. Alienation of labour means that people become detached from their own activities; that is, they create a distance between themselves and their creative and productive capacities. In his work on Marx, psychologist Eric Fromm [22] writes: “For Marx, the process of alienation is found in work and the division of labour. Work is an active connection between man and nature, a process of creating a new world, including man himself… But with the development of private property and the division of labour, labour loses its character as an expression of human agency; labour and its products become separated from man, from his will and his plans.” Humans understand and shape their identities through the activities in which they engage and the inner forces and drives they invest in them. By creating and shaping the world and its environment, and by focusing their energies on external projects and goals, a person realises a sense of self in the world. Activity, rather than passive consumption or tasks, defines what Marx called human nature. As Fromm pointed out in his book To Have or To Be, it is through "being," not "having," that one experiences meaning and purpose in life. It is through "being in love," "being passionately," "being active," "being hopefully," "being purposefully," and "being productively" that one can live a fulfilling life without having to rely on "having" (love, passion, positivity, etc.). The division of labor emphasizes "having," and many of the jobs and activities involved do not encourage people to develop a sense of "being." As the German poet Johann Wolfgang von Goethe said, humans strive to exist by "transforming themselves from the night of possibilities into the day of actualities." Humans seek to become themselves by externalizing their internal values ​​and assumptions, and this process is achieved through activity. Fromm wrote: "Man is alive only in productive action, only in so far as he grasps the external world by the expression of his specifically human capacities and grasps the world with these capacities... In this productive process man realizes his essence." The division of labor reduces the sense of human agency, that is, the ability of humans to actively influence and direct their own lives in meaningful ways that they value, choose, and exercise their will.

By creating a "qualitatively rich environment", the hacker paradigm and its promotion of open source provide multiple "qualitatively rich opportunities" that promote the progress of our social and political structures. It has the potential to realize the more positive aspects of human nature - human agency - rather than just its negative and consumptive parts. The main assumption here is that if artificial intelligence is to develop and integrate into the economic base, it should make people's lives better. More specifically, if automation occurs, the result should be to improve people's lives, not to harm them. One way to achieve this goal is to ensure that "qualitatively rich opportunities" are distributed enough so that everyone can benefit from them, and open source is a way to achieve this goal. One important reason why the development of artificial intelligence may lead to people's worse off and suppress their potential opportunities is the existence of monopolies that use proprietary models to control artificial intelligence and dominate its market use. The hacker paradigm and open source provide possible solutions to avoid this outcome.

Section 2.2

Before understanding how monopolies reduce "qualitatively rich opportunities", we need to further clarify what "qualitatively rich opportunities" are and what capabilities they inspire and help develop. After explaining the nature and significance of such opportunities, we will explore how monopolies significantly reduce the accessibility of these opportunities and what this means.

The idea of ​​"qualitatively rich opportunities" originated from Aristotle [23] and was further developed by philosopher Martha Nussbaum and economist Amartya Sen. Aristotle believed that humans must achieve certain functions or ways in order to flourish and live a full and dignified life. He believed that there was a "human nature" that contained both basic and higher-order needs. Regardless of historical and accidental changes, humans have certain fundamental characteristics. Of course, these characteristics are reflected in historical contexts, but certain essential consistencies can be observed in all cultures and civilizations. Aristotle said: "The good is the goal of all things." The so-called "good" is the proper function of a thing. For example, the good of the eye is to "see" and the good of the seed is to "grow into a tree." To achieve the proper function of something is to achieve its "flourish", that is, to achieve the state it pursues and strives towards. To achieve such a state is to gain an intrinsic value.

To achieve its functions as a human being, you need basic necessities of life, such as food, water, shelter, and the ability to move. But these goods themselves are not intrinsically valuable; they are valuable because they enable people to function in pursuit of further, more intrinsically valuable things. If a person has only these basic conditions for survival, we can say that he is "functioning", but not necessarily "functioning well". For a person to "function well", he must have other resources of higher intrinsic value. Humans need education and access to information to develop their cognitive abilities; they need family and close relationships to develop social skills, personal identity and emotional processing; and they need channels to exercise their will, explore their interests, engage in passionate pursuits, and cultivate a sense of purpose. A good political and social arrangement, the kind of social system we should pursue, should provide people with specific opportunities to exercise these needs and thus achieve good human functioning. "Qualitatively rich opportunities" are opportunities that provide the basis for people to develop intrinsically valuable abilities (such as the ability to make informed decisions, the ability to set worthy goals and the paths to achieve them, and the ability to establish meaningful and influential interpersonal relationships). Social institutions such as work, education, community and the arts are qualitatively rich environments that stimulate and promote the abilities needed to achieve these goals.

Qualitatively enriching opportunities are important because they enable people to develop “agency.” Borrowing a concept from Stanford University psychologist Albert Bandura [24], we can understand “human agency” as an individual’s ability to feel that they have control and influence over their lives. More specifically, qualitatively enriching opportunities can stimulate behaviors that target “self-efficacy.” Self-efficacy refers to a person’s sense of confidence in the face of challenges and obstacles. When faced with a difficult and novel task, people with high self-efficacy assess themselves as having the ability to overcome difficulties and complete the task; they are motivated to persevere and overcome failure; and they are able to maintain a good psychological state even in unfamiliar and complex situations. Qualitatively enriching opportunities are designed to develop the abilities that enhance self-efficacy and are needed to cope with specific situations. Through these opportunities, a person is able to exercise his or her agency, gain control, and thus expand his or her sense of freedom. As Bandura (1994) [25] puts it, “The stronger a person’s sense of self-efficacy, the more challenging the goals he or she sets for himself or herself and the more committed he or she is to those goals.” Therefore, by enhancing self-efficacy, individuals will gain a higher degree of freedom, both at the level of belief and practice, and live a more meaningful and richer life.

Work plays a vital role in shaping individual identity and self-identity. It provides people with the opportunity to explore different career paths and enable them to discover what they want to pursue in life. Work is often an important way to build professional and personal relationships and integrate into different social networks. It is also a medium for people to acquire capabilities and climb the social ladder. The work domain can also help people develop motivation, intelligence, and attitudes in a variety of environments and situations. In other words, work provides people with the opportunity to develop certain functions or lifestyles that are essential to achieving a good life.

For example, imagine a young person entering the workforce for the first time. He tries a few jobs. He finds that he doesn’t like retail because of the constant interaction with customers. He tries delivery service, but finds that he doesn’t like the repetitive, day-to-day work. Finally, he finds work on a construction site and finds some of the tasks interesting and challenging. He likes the feeling of working with his hands and decides to pursue a career in the trades. This person then attends technical school, where he builds important relationships, is challenged in a new environment, and is inspired to improve his skills, accumulate knowledge and experience, and constantly face and overcome new difficulties and obstacles in the process. After graduation, he seizes an opportunity—he is given a lot of attention and support by a mentor who appreciates his work ethic. He then works for a company for a while and builds enough knowledge and confidence in the process to decide to start his own company to achieve this goal. He begins to build customer relationships, focuses on skills that will set him apart from the competition, and eventually builds a successful business.

This may sound idealistic, but it is a familiar and realistic example. It depicts a life of purpose, formed through opportunities found in environments and situations that provide experiences that generate a sense of agency and self-worth. Of course, a range of other conditions are usually required to achieve this picture of life, such as growing up in a supportive environment and having the resources and time to explore personal interests - but in the United States, work is undoubtedly the core channel for developing the key functions that constitute a good life. The person in the above article achieved success and achievement by learning new skills, overcoming challenges, and making critical choices to cope with social environments. As he grew in his professional field, he continued to set new goals and gained enough self-efficacy to believe that he could achieve them. He developed a complete set of capabilities and ultimately gained a sense of dignity. Regardless of whether this person has other good life experiences in the past or in the future, the quality of his life will be greatly reduced without the opportunities that allowed him to work and grow.

Here is a real example, but the person will remain anonymous. After graduating from university in the Netherlands, he didn’t want to get a “normal job” and began traveling to other countries. He made some money making music on YouTube and stayed in hotels and hostels. But as his money dwindled, he began to feel lonely and isolated, and became depressed. He had complete freedom—he could go anywhere, see anything, do anything he wanted—but he felt extremely lonely. Lying alone in a bed in a hostel in Asia, staring at the ceiling, with no money, no girlfriend, and no job, he remembered what his father had said to him: “He taught me when I was a kid: whenever I felt frustrated or didn’t know what to do, I would go find a pile of sand, get a shovel, and move the sand from one side of the yard to the other… Do some physical work, do something hard—do something. I did the same with my entrepreneurial projects.” So he began learning to code, developing software and apps, and using the Stripe platform [26] to publish his work to the public to raise funds. He focused on activities that gave him a sense of direction and creative opportunities. In true “hacker spirit”, he saw the world as a fascinating field full of problems to be solved, some of which he thought were tractable and worthwhile, so he tried to solve them. Amazingly, he set a goal: to start twelve startups in a year - and he actually did it.

This person initially felt alienated and lonely, lacking a purpose in life. So he chose to take action and threw himself into creative expression using basic programming languages ​​and tools. By leveraging artificial intelligence, he was also more able to create valuable and practical products for others to use and experience. He took his father's advice and threw himself into practical actions, and as a result, his depression and anxiety gradually eased. In an information-rich environment, he found many "qualitatively rich opportunities" to develop himself and improve his self-efficacy and ability. He connected with more challenging goals and continued to work hard to achieve them. And open source technology played a key role in this process.

This is also the significance of the Dora ecosystem. Through platforms like DoraHacks.io, people like the ones described above can collaborate with developers around the world through "qualitatively rich opportunities" to innovate together and create meaningful solutions that people find urgent and relevant. The ecosystem provides developers with the necessary resources to join "BUIDL" teams anywhere in the world to solve fascinating problems; feel a sense of community with other builders in hackathons; and drive them through incentives to achieve meaningful goals, which often lead to more challenging and stimulating next-stage tasks. The Dora ecosystem creates an environment full of possibilities, enabling people to build decentralized products and services, such as returning ownership to users themselves, increasing their autonomy, security, and control over financial assets in the digital world. As discussed further below, large companies and monopolies do not usually benefit directly from these types of technologies. So what is their motivation to fund such technologies? The open source technology stack can improve the quality of value because it provides people with a path to develop more meaningful products, which are often more meaningful than those produced by large, narrow-minded institutions.

Therefore, if automation has the potential to reduce or even eliminate the “qualitatively rich opportunities” for people to exercise their agency and gain a sense of self-efficacy, then the space it vacates should be filled by other qualitatively richer opportunities. If there is such a thing as “human nature”, and this human nature includes a range of abilities that need to be developed to enable people to live a dignified life, then the best and most desirable political and social arrangements should be to distribute “qualitatively rich opportunities” fairly to ensure that people have the choice to exercise their abilities and become individuals who can function well and achieve a prosperous life. Providing people with qualitatively rich opportunities so that they can use them and organize, mobilize and focus their inner strength on practices that can exercise and inspire individual abilities constitutes a foundation for people to live a more meaningful life. It is giving people the freedom to become the people they value.

The next section will turn to a discussion of AI monopolies to reveal how they suppress such future opportunities.

Part 3

Section 3.1

The potential monopolists in the field of artificial intelligence are self-evident: Google, Apple, Microsoft, Meta, and Amazon. These companies and a few others account for 65% of fixed web traffic and 68% of mobile web traffic in 2024 [27]. Google accounts for about 90% of the global search market, and its brand name has almost become synonymous with the act of "search" itself; Meta owns major social media platforms such as Instagram, Facebook, and WhatsApp, which are highly visited platforms for people's daily communication and information sharing; Microsoft controls about 72% of the desktop operating system market; in the United States, Apple accounts for about 60% of the smartphone market share; and Amazon dominates online retail, with a scale "greater than the combined size of the next 15 largest e-commerce retailers in the United States" [28]. Moreover, Google, Meta, and Amazon together account for about 60% of advertising revenue in the United States. The products, beautiful ads, and attention we encounter online are driven primarily by these large technology companies. According to Synergy Research [29], by the end of this century, Microsoft, Google, and Amazon alone could control up to two-thirds of the world’s data. These companies also control much of the cloud infrastructure—Amazon, Microsoft, Google (and Alibaba) together have 67% of the global cloud computing market share [30]. These companies have vast reserves of resources and have a long history of using a variety of means to exclude competitors, including eliminating potential threats through mergers and acquisitions, using market dominance to promote their own products first, practicing predatory pricing, exploiting network effects, and other well-known strategies. For example, given Amazon’s market dominance in e-commerce, it can prioritize its own products over those of other suppliers that may be of higher quality. Consumers are often more inclined to buy Amazon products when they are offered at a lower price, and Amazon has the resources to maintain this low-price strategy. In terms of network effects, “most tech giants now have billions of users [31],” creating an almost insurmountable barrier for startups and smaller companies. Google has also been accused of deliberately down-ranking certain content and prioritizing information that favors its own interests.

Today, these companies are actively competing for dominance in the development of artificial intelligence. “Meta, Microsoft, Amazon, and Alphabet are expected to invest a cumulative $325 billion in capital expenditures and investments in 2025… This represents a 46% increase from the approximately $223 billion reported by these companies in 2024 [32].” Moreover, “almost without exception, all startups, new entrants, and even AI research labs rely on these large tech companies: they rely on the computing infrastructure of Microsoft, Amazon, and Google to train their systems, and on their vast consumer market channels to deploy and sell AI products [33].” Tech giants have monopolized the key resources that startups rely on for survival [34]—talent, data, and computing power.

Existing industry giants are developing various models to maintain their foothold in their respective markets. Because they control most resources—for example, “these companies contribute more than 22% of the market value of S&P 500 companies, and their individual market value even exceeds the GDP of some G7 countries such as Canada and Italy [35]”—they can squeeze out competitors at will and prevent startups and small companies from entering the market. As mentioned earlier, as DeepSeek continues to make progress as a competitor in the global market, these largest AI companies are hyping up an “arms race” and exaggerating the Chinese threat and the potential harm of artificial intelligence, thereby promoting regulation and development restrictions on AI, which are ultimately most likely to benefit themselves. At the same time, because the monopoly system relies on proprietary models whose architecture, weights, learning algorithms, code, and embedded goals are hidden from the public, the economic and market dominance of these companies will also shape people’s thinking and behavior in a “hidden way” as AI infiltrates the information network.

So, can regulation solve this problem? MIT Technology Review recently noted that “regulation can help, [36] but government policy often ends up reinforcing rather than weakening the power of these companies—they are able to use their capital and political influence to manipulate policy.” Alex Rampell [37] recently wrote [38] that “the Biden administration’s executive order [39] seeks to impose artificial limits on computing power and ban open source technology on the grounds that it poses a threat to national security, while effectively opening the door for a few of the largest companies to monopolize regulatory resources.” For example, although Sam Altman denied this in a Free Press podcast with Barry Weiss [40], investor Marc Andreessen has repeatedly warned that the real purpose of Biden’s AI executive order is to put control of AI in the hands of a few companies (most likely OpenAI) and eliminate competitors by raising barriers to entry. Furthermore, although this move is controversial and perhaps even naive, since DeepSeek released its V1 and V3 models, Anthropic CEO Dario Amodei has been calling for stricter export control policies [41] to further restrict China’s ability to obtain chips. However, as Dylan Patel [42] and Nathan Lambert [43] pointed out in the Lex Friedman Blog, DeepSeek’s open source model does not obviously pose any substantive national security threat. But it does pose a threat to ecosystems that rely on proprietary systems and control potential value through models. DeepSeek’s approach forces this “proprietary capital” to move toward public resources that can be used and built by anyone. As such, Amodei’s insistence on a “unipolar vs. bipolar world” actually contains competitive motivations: either AI power is concentrated in the United States, or both China and the United States have equal AI capabilities—the latter, in Amodei’s view, must be avoided.

Perhaps we should adopt a less hawkish optimism. Economist Tyler Cowen has made a strong case for finding a path to development through cooperation[44] and win-win situations rather than an all-out arms race. Competition is still necessary, of course, but the competitive mentality that drives future events should not be based solely on Darwinian logic of “survival of the fittest” but should be closer to the ancient Greek tradition of “Agon”[46] observed by Nietzsche[45]—a social system in which artists, military leaders, and athletes encouraged each other and grew together through competition.

Without opponents, competitors will be lost. Without the opportunity to improve, break through, and innovate, individuals lose the possibility of becoming better and more creative. The very existence of competitors depends on people willing to confront them. It was precisely because of challenges that Greek poets were able to imagine more deeply and express more brilliantly. Nietzsche pointed out that the greatest danger facing Greek athletes was not failure but the loss of their competitors. “This is the very heart of the Greek athletic idea,” he writes: “its aversion to dictatorship, its fear of its dangers, its desire to counteract the overwhelming power of one genius with ‘another genius’.” We should not understand competition as a struggle for survival in the state of nature, a Hobbesian situation in which resources are scarce and people threaten each other—“lonely, poor, nasty, brutal, and short.”[47] Nor should we see the competitive instinct as a remnant of some primitive sexual drive to suppress rivals and eliminate threats. We should understand it as an instinct that can be ennobled, as Nietzsche put it, “the only fertile soil for all human emotions, actions, and works.” The competitive instinct itself is virtuous and should be cultivated as something that works “at the right time, in the right way, and for the right reason” (Aristotle, Nicomachean Ethics, Book II), rather than being seen as a necessary evil. Amrita’s fear of China may not be the right response—perhaps a better, more virtuous response would be to see China as a worthy competitor.

Section 3.2

Let us now focus on why these potential AI monopolies, by dominating the AI ​​ecosystem, could suppress the wider diffusion of qualitatively enriching opportunities. There are at least two reasons: the first concerns economic agency, and the second concerns value agency. Economic agency is the ability of people to enter the market and create value and utility that meet consumer needs and desires. If the market for AI models is concentrated in the hands of a few companies, the opportunities for emerging companies and entrepreneurs to make valuable contributions to the world will be limited. Innovations that respond to democracy, health care, financial autonomy, political rights and freedoms, or other major and meaningful issues will be greatly reduced, thereby reducing the social benefits that AI can bring. Any products and services that threaten the control, influence, or wealth of these few AI monopolies will be directly excluded from the market, rather than determined by the natural mechanisms of the market or the real needs of consumers.

Value agency, on the other hand, refers to people’s ability to influence and have a say in what society should pursue. If AI is trained under opaque conditions, based on values ​​and goals that are not accessible to the public and cannot be engaged, then people’s influence on these issues will be weakened. If proprietary models determine the nature of most or almost all products and services embedded with AI, then people’s control over these technologies will be greatly reduced. For example, the so-called “alignment problem” refers to how to align AI with human interests. So, what are human interests? What are the goals that constitute a good and worthwhile society? What moral and ethical principles should AI operate according to? Can these values ​​be determined by a few companies without making their training data and learning algorithms public? Can they be free from public scrutiny and engagement? If the technical architecture is closed off from public view, then the answers to the above questions are more likely to come from authoritative decisions rather than consensus obtained through democratic processes.

The dual loss of economic agency and value agency is dangerous, and the rest of this article will further explain why.

Section 3.3: Economic Activism

As Austrian economists Ludwig von Mises and Friedrich Hayek famously pointed out, the Soviet Union was destructive in some ways in part because it could not allocate resources properly through naturally occurring market mechanisms. Under such a system, internal central planners determined the needs and desires of the people - and the information they had was scarce, biased, and inaccurate, making it impossible to accurately allocate them. Therefore, a large amount of bureaucratic and coercive means were implemented to allocate state resources.

For example, the state had to set prices. At least in the Soviet case, there was no widely available free market where goods and services were regulated by supply and demand. Producers produced according to state commands rather than oriented themselves according to market incentives (such as competition) - which should coordinate production and consumption and promote market equilibrium. In other words, prices lose their key informational function—they are supposed to indicate where resources should flow. “Any control over the price or quantity of some commodity deprives competition of the power to coordinate individual efforts, because price changes do not reflect relevant changes in the environment and provide no reliable guide to individual action,” Hayek wrote in The Road to Serfdom.

But more importantly for this article: state-controlled production and economic planning also means that people are deprived of the opportunity to enter the market and create value, making it difficult for them to improve their lives through innovation. Any innovation that might threaten state power, weaken its legitimacy and influence, or transfer some of its control will be prohibited from entering the market. For example: suppose someone observes that in remote rural areas of a country, people cannot build roads due to frequent flooding, making it difficult for them to participate in national elections. The inventor develops a new type of concrete that is not corroded by floods and hopes to build roads in the countryside to improve people's access and make it easier for them to participate in public affairs. But when he applies for resources from the National Transportation Commission, he is rejected. The reason is that the region’s residents generally want to have autonomy over agricultural production, so the state is unwilling to provide these people with access to the electoral system. In the absence of market competition, there are no other opportunities and mechanisms to build these paths.

Monopoly, especially when protected by a regulatory system, poses a similar threat. Marx once made a famous criticism: as a social system, the inherent logic of capitalism will gradually and inevitably lead to the concentration of the means of production in the hands of a few. Although many economists, including Hayek and Mises, believe that this view is wrong and believe that this result is avoidable in a properly functioning capitalist system. But if artificial intelligence is monopolized, then Marx’s prediction will be closer to reality than ever before—whether this is due to historical laws or unforeseen structural variables. As Chris Dixon [48] observes in his book Read, Write, Own, start-ups have found space in the market in part because large companies are often short-sighted and fail to detect in time that emerging products are gradually gaining attention from consumers. In the process of focusing on their own business, these giants often miss market trends and movements, and if startups are lucky enough, they can accumulate enough momentum to become competitive. But if these old giants control artificial intelligence, they may use AI's capabilities to eliminate the advantages that startups could have taken advantage of. Over time, monopolies will have an impenetrable control over the market, and the value chain of various industries will eventually be further "upstream concentrated" in the hands of these companies - even more than it is now. Disruptive technologies that are ignored by experts but can create unexpected value in the market may have their innovation space completely squeezed out.

Because of its intelligent capabilities, artificial intelligence has great potential as a capital tool to create advanced technologies that improve human life. However, because some of the most beneficial applications for the public may not be in the interests of governments and large companies, its potential benefits may be tragically weakened in the power structure. The most obvious example is financial autonomy.

Look at the case of Bitcoin to illustrate the problem. Satoshi [49] invented Bitcoin to give people the ability to have financial freedom—by creating a scarce resource that can store value similar to gold. Bitcoin enables people living in countries with high inflation and extremely devalued currencies to accumulate wealth that can be transferred globally. People can own an asset that will not depreciate due to excessive money supply or rising national fiscal deficits. Bitcoin can help people live more stable and secure lives by providing transparency, personal ownership, and institutional arrangements that incentivize social cooperation—all without relying on compulsory trust or moral charity.

One ​​would hope that those in power would find ways to incorporate the enormous value that Satoshi has brought to the world into the existing system. This is not to say that Bitcoin should immediately replace fiat currency, but given the promising solutions to the potentially fatal problems of the fiat currency system, governments should at least make greater efforts to incorporate these solutions into practice. It is reasonable to expect that those in power will find a way to integrate blockchain, Bitcoin, and their potential to enhance economic freedom. Until recently, however, regulatory progress has been slow and confusing. Under the Biden administration, a lack of clarity and specificity has prevented many initiatives around cryptocurrencies from realizing their potential and hindered their more efficient and productive use.

Technologies that have the potential to improve human life could be easily abandoned if AI is monopolized through regulatory frameworks. Most people will not be able to use these proprietary models through licensing or patents if they conflict with government or corporate interests, and small and medium-sized companies will be absorbed by mergers and acquisitions. Although AI could revolutionize election platforms and the way people participate in elections; although it could enhance individual freedom through better data security or the creation of more economic opportunities, the path to explore these possibilities is likely to be blocked if it is inconsistent with the interests of the companies that control most (or even all) of the AI ​​technology stack. If a potential AI technology has the potential to threaten advertising revenue by solving a problem that some industries rely on for profit, or even parasitic problems, then technology companies like Google and Meta will almost certainly block attempts to bring the technology to market.

One ​​of the Trump administration’s most significant policy initiatives to date has been its new cryptocurrency policy. The policy repeals the previous administration’s executive order and seeks to develop a regulatory framework that will help advance the technology. This has important implications for other forms of innovation taking place in the field of artificial intelligence. With alternative sources of funding to traditional venture capital, such as tokenomics [50], startups can fund their projects and build momentum.

In his book Zero to One, Peter Thiel argues that startups are and should be seeking to become monopolies. They need to occupy a specific space in the market, beat out other competitors by developing proprietary technology, and further exclude new entrants from entering the market. As Thiel argues, this strategy is critical if companies do not want to be losers. If they want to gain the attention of investors, they must take the path of establishing a monopoly. However, with the help of tokenomics and other mechanisms related to cryptocurrency, people also have the opportunity to develop valuable technology while retaining the spirit of open source. Under such a mechanism, cooperation and collaboration can be promoted, and the funds needed to support growth can also be obtained.

Section 3.4 Value Mobility

The reason why economic mobility is weakened is that the opportunities for quality value creation in the market are decreasing, making it difficult for individuals to meet changing needs and desires through innovation and production. When the AI ​​technology stack is controlled by a few monopolies and dominates the entire ecosystem, even if someone identifies a truly meaningful problem and designs a powerful solution, if it is not in the interests of large companies, the solution is likely to not enter the market. Economic activities and the social value they bring should not be based on an authoritative mechanism that allows power to arbitrarily suppress and stifle things that are beneficial to society as a whole. So how do AI monopolies compress people's "qualitatively rich opportunities" in value mobility?

If Ray Kurzweil’s Singularity theory and its associated predictions are even partially true—if humans are indeed going to merge with nonbiological intelligence and live in an increasingly artificial environment—then the monopoly of artificial intelligence is not only undesirable, but also arguably dystopian. This vision evokes images of “technological overlords” who dominate human society, while people either become slaves to intelligent machines or comfortably and contentedly enjoy the material fruits of intelligent prosperity. In this picture, someone will always have to mine mineral resources, while others will consume the abundant goods and services produced by them. But if we assume that AI will develop quickly enough for intelligent robots to do the work of mining without human intervention, then everyone will eventually become a pure “consumer.” “In the next few decades,” Kurzweil predicts, “almost all routine physical and mental activities will be automated.” At that time, only those jobs that are social, creative, innovative, or highly unpredictable will be open to people with the relevant skills and training, and the rest will mostly become consumers. With the help of nanorobots, virtual reality, non-biological intelligence enhancement technology and ubiquitous information flow, our reality will be closer to what philosopher Robert Nozick envisioned as the “experience machine”. Through the intervention of some device - whether it is connected to a simulation device or fully immersed in a digital and information environment like virtual reality (VR) - people can experience anything they want. Any sensory experience, emotion, thought or event will be available to anyone at any time and available for consumption. One can experience the profound satisfaction of writing the “Great American Novel”, bravely confronting the jury in an Athenian court like Socrates [51], or bravely standing on a verdant hill overlooking the enemy like Napoleon, or climbing Mount Everest in a simulated climb and experiencing the glory of perseverance and the view from the top of the world. What matters is not whether these events actually happen, but that someone “desires” it, and it will “exist”. In this vision, the goal of life is consumption; and the "moral life" - the life worth pursuing - is defined as the pursuit of internal stimulation. The external world, the individual's true state of existence will no longer matter. Whether a person is truly brave, compassionate, loving, just, pious, good or evil - this no longer matters. People's interests and conflicts will be reconciled in the consumption of artificial experiences. In a world where "problems have been solved", there are no specific problems that need to be solved by actors, and actors become dispensable; the only thing that is needed is "consumption".

The real danger of this situation is that monopolies will have complete control over the algorithms that generate technology, that is, control the information structure we have access to. Information will continue to be embedded in a consumption-centered environment. Ideally, people should have the freedom to do what they want and live the way they agree. But the way people "see" the world, the way they construct reality, the social context they live in - will be filtered and shaped by those companies that have integrated intellectual capital and become the sole producers of goods and services in the future. Philosopher Allan Bloom wrote in his book The Closing of the American Mind: “Every educational system has a moral purpose it hopes to achieve, and structures its curriculum accordingly. It hopes to shape a certain type of person.” AI monopolies will shape the meaning of “what it means to be human” by transforming the human environment to suit their own interests. And the things that really matter—the opportunity to exercise agency, to achieve a true sense of self-efficacy, to discover what you truly value and desire in life, to construct life goals that make reality more consistent, livable, and unified—will be strongly influenced and manipulated by corporate interests. And these interests are essentially built around economic growth and profit maximization.

In their book Manufacturing Consent, [54] Noam Chomsky [52] and Edward Herman [53] argued that the American media was in fact a propaganda tool rather than an objective source of information (a view that was controversial at the time but is now mainstream). They argued that this propaganda function operated through a set of “filtering models” that took information as input and produced perfectly standardized outputs for the public’s “good.” Some of these media filtering mechanisms are precisely what the tech giants use today. The first filter is the size of media companies, their ownership structure, and their profit-oriented business model. This makes the investment and entry costs required to enter the media market prohibitively high, making it difficult for other competitors to survive. The legal system is used to enforce a set of standards that suppress startups.