Musk vs. Argentina

Here is a thought experiment that will make this concrete.

Imagine a war between Elon Musk and Argentina. Argentina is the 32nd most powerful military in the world according to the 2026 Global Firepower Index. It has 46 million people, an active military of over 80,000 troops, a navy with submarines and destroyers, an air force with fighter jets, and a GDP of roughly $640 billion. It fought Britain over the Falkland Islands. It is a real country with a real military.

Musk’s net worth, approximately $800 billion, exceeds Argentina’s entire GDP. But wealth is the least interesting dimension of the comparison. Consider the actual instruments of power.

Orbital infrastructure. Musk controls over 10,000 Starlink satellites in low Earth orbit, operating in 115 countries. He demonstrated in Ukraine that he can unilaterally decide whether a country’s military communications work or don’t. Foreign Policy called it the “God Switch.” Argentina has zero military communication satellites. Its armed forces depend on commercial providers that Musk either owns or can pressure.

Defence technology. SpaceX holds over $6 billion in Pentagon contracts and is part of the Golden Dome missile defence programme. Through his network (Peter Thiel’s Palantir, Palmer Luckey’s Anduril) Musk sits at the centre of a private defence-tech constellation that is displacing legacy contractors. Argentina’s military hardware is largely ageing Western and Israeli equipment it cannot independently maintain.

Information warfare. Musk owns X, a platform with over 600 million users. Its algorithm is powered by Grok, an AI system he also controls, now integrated into US Department of Defence classified networks. He can shape global narratives about any country in real time. Argentina’s state media barely reaches beyond its borders.

Economic leverage. With hundreds of billions in liquid and semi-liquid assets, Musk could fund opposition movements, wage sustained economic pressure campaigns, and outlast Argentina’s treasury in any war of attrition. Argentina’s foreign reserves sit at roughly $30 billion.

Cyber and AI capabilities. Musk’s companies employ some of the most capable engineers on Earth. xAI, Tesla’s autonomy division, and SpaceX’s orbital mechanics teams represent a concentration of technical talent that exceeds the entire technology sector of most nations.

Argentina retains one critical advantage: conventional military force. Musk has no standing army, no navy, no air force in the traditional sense. He cannot physically invade Buenos Aires. But this objection betrays a 20th-century understanding of warfare. In the 21st century, the country that controls communications, shapes information, denies orbital assets, and wages economic war doesn’t need to invade. The invasion is the old way. The new way is to make the other side unable to function.

Musk would win. And it would not be close.

Now consider the implications. Argentina is ranked 32nd. Every country ranked below it, roughly 113 nations, would face a similar or worse mismatch.

This is not a theoretical future. It is the present.

Why This Is Happening

The Logic of Violence

In 1997, James Dale Davidson and Lord William Rees-Mogg published The Sovereign Individual, a book that has become something of a sacred text in Silicon Valley. Peter Thiel calls it the most influential book he has ever read. He wrote the preface to its 2020 reprint.

The book’s core insight is simple. Political structures are not determined by ideology. They are determined by the economics of violence. Whoever can most efficiently project coercive force sets the terms for everyone else.

Davidson and Rees-Mogg introduced the concept of “megapolitics,” the study of how the cost structure of violence shapes governance. They traced four stages:

In the Agricultural Age, heavy cavalry and fortified castles created aristocratic warfare with enormous fixed costs. Power concentrated in landed lords. The result was feudalism.

In the Gunpowder Age, cannons destroyed castles and musket armies could be trained quickly from the peasant population. This created increasing returns to scale in violence. Bigger armies always won. The nation-state was born, not as a philosophical choice, but as the only political form that matched the military-economic reality. Every feature of the modern state (taxation, standing armies, nationalism, welfare systems) follows from this underlying dynamic.

In the Information Age, the economics reverse. Digital defence is cheap relative to digital attack. Mobile knowledge workers can flee taxation. Encrypted money escapes confiscation. The gap between offensive and defensive capability that made nation-states dominant is closing. As coercion becomes more expensive relative to the value it can extract, states lose their structural advantage.

What they got right is striking. In 1997, they described with mechanical precision a system of cryptographic digital money based on prime numbers, divisible, pseudonymous, operating outside central banks. Bitcoin arrived twelve years later. They predicted remote work, the gig economy, location-independent knowledge workers, populist nationalist backlash, cyberwarfare, and jurisdictional tax competition (later confirmed by the Panama Papers). The directional accuracy is remarkable.

What they got wrong is equally instructive. Nation-states have not collapsed. Government revenue as a percentage of GDP has barely changed across the OECD. The secessionist movements they predicted (Quebec, Catalonia, Scotland) have stalled. Cyberwarfare expertise scaled with state resources, not individual resources. The most sophisticated cyber actors remain the NSA and China’s PLA, not lone hackers. Their timeline was wildly optimistic.

But the trend they identified is real. Even if the pace was wrong and the endpoint is not the libertarian utopia they imagined.

The Entertainment Trap

There is a second intellectual pillar worth introducing. In 1985, Neil Postman published Amusing Ourselves to Death, in which he argued that Aldous Huxley’s dystopia, not George Orwell’s, was coming true. The threat was not a government banning books. It was a population that no longer wanted to read them.

Postman’s insight explains the demand side of the Great Houses thesis. Even if the structural conditions for challenging state power exist, someone must resist. Postman explains why no one does. A population trained by algorithmic feeds to seek entertainment above truth, to confuse outrage with participation, to evaluate politicians as performers rather than stewards, cannot hold power accountable. The Great Houses don’t need to overthrow the state. They just need the populace to be too entertained to notice the transfer of power.

Shoshana Zuboff’s Surveillance Capitalism, Jonathan Haidt’s The Anxious Generation, and Tim Wu’s The Attention Merchants have all confirmed and extended Postman’s framework into the age of TikTok and algorithmic manipulation. The picture they paint is consistent: a population that has been trained to seek entertainment above truth, that cannot sustain attention, that confuses engagement with participation, and that voluntarily occupies media environments designed to exploit its psychological vulnerabilities, is a population that cannot hold power accountable.

The population is not being repressed. It is choosing its captivity. And that is the most effective form of control ever devised.

The Changing Logic of Violence

The economics underpinning state power are inverting in real time.

The Cost Asymmetry Revolution

A 22-year-old with a $500 FPV drone and a gaming controller can destroy a $3 million tank. The ratio is 6,000:1 in the attacker’s favour. The US Army is now actively seeking a new interceptor under $1 million specifically because the current economics are untenable.

The Historical Parallel

This has happened before. The crossbow was banned by the Second Lateran Council in 1139. Not because it was inaccurate, but because it was democratising. A peasant with a crossbow could kill a knight who had trained for twenty years and whose equipment cost the equivalent of a small estate. The weapon undermined the social logic of feudalism (that military power and social status were inseparable) before gunpowder finished the job.

Gunpowder then destroyed the castle, which destroyed the feudal lord, which created the nation-state. We are watching the equivalent process unfold again. Cheap, mass-producible offensive technology is inverting the cost advantage of large, well-funded defenders. The drone is the crossbow of our age. The question is what new political form emerges from the rubble.

Private Companies as Military Actors

This is not only about drones. Private companies now exercise functions that were exclusively sovereign within living memory.

SpaceX/Starlink became the backbone of Ukrainian battlefield communications after Russia destroyed terrestrial networks. Musk personally declined to enable Starlink coverage near Crimea to prevent a Ukrainian drone offensive against the Russian fleet. A private citizen overriding a sovereign nation’s military planning in an active war. In May 2026, he publicly accused the US military of misusing civilian Starlink for drone operations against Iran. A private CEO was constraining American military operations.

Palantir’s Gotham platform is used by the US intelligence community for target identification, autonomous sensor tasking, and kill-chain acceleration. The state’s analytical brain increasingly runs on private software.

Anduril Industries, founded by Oculus creator Palmer Luckey and backed by Peter Thiel’s Founders Fund (which wrote its largest single cheque, $1 billion), beat Boeing, Lockheed, and Northrop Grumman for the Air Force’s Collaborative Combat Aircraft programme. It holds a $20 billion Army framework contract for AI counter-drone systems. A company that did not exist in 2016 is now displacing the military-industrial complex that won the Cold War.

This too has historical precedent. The British East India Company maintained a private army of 260,000 soldiers at its peak. Twice the size of the British Army. It governed 90 million people, collected taxes, operated courts, minted currency, and conducted foreign policy. It was, for two centuries, the most powerful private organisation in human history.

The Dutch East India Company commanded 50,000 soldiers and 160 warships. It waged wars, signed treaties, and colonised entire territories as a profit-seeking corporation.

The condottieri of Renaissance Italy, private military contractors hired by the great merchant houses, literally seized states. Francesco Sforza, a mercenary captain, made himself Duke of Milan. The Medici bankrolled armies and used them to maintain political control over Florence for three centuries.

And the Knights Templar, a private military-financial corporation founded in 1119, grew so powerful that the King of France had to destroy them.

When private entities accumulate military capability, they accumulate political power. The East India Company, the VOC, the condottieri, the Templars. Each eventually rivalled or exceeded the sovereign authority they nominally served. SpaceX, Palantir, and Anduril are the 21st-century heirs to this tradition.

The Great Houses: A Definition and a Portrait

I use the term “Great Houses” deliberately. Borrowed from both history (the great feudal houses of medieval Europe) and fiction (Frank Herbert’s Dune, where noble houses controlled entire planets and waged wars for resources), the term describes something specific. Ultra-wealthy individuals or dynasties who have accumulated enough power, technological, financial, military, informational, to rival or exceed the capabilities of nation-states. Not merely rich people. People who control the instruments of sovereignty: communications infrastructure, defence technology, information architecture, orbital systems, and AI.

Here are the ones that matter most today.

The House of Musk

The most powerful Great House in history. Net worth approximately $800 billion, exceeding the GDP of all but about 25 countries. But the wealth is the least of it.

Musk controls Starlink (10,000+ satellites, the communications backbone of active wars), SpaceX (the Pentagon’s dominant launch provider, integrated into missile defence), X (600 million users, algorithm powered by his own AI), xAI (Grok, now integrated into DoD classified and unclassified networks), Tesla (manufacturing, AI, robotics, targeting 10 million humanoid robots per year by 2027), Neuralink (brain-computer interfaces), and The Boring Company (infrastructure).

No individual in history has simultaneously controlled orbital infrastructure, a global media platform, an AI laboratory, a defence contractor, and held a direct role in government. This is not influence. It is a vertical stack of sovereign capability.

The House of Thiel

Peter Thiel is the intellectual architect of the movement. Co-founder of PayPal and Palantir, early Facebook investor, net worth approximately $15 billion.

In 2009 he wrote: “I no longer believe that freedom and democracy are compatible.” He named The Sovereign Individual the most influential book he ever read and wrote its 2020 preface. His protege, JD Vance, is the Vice President of the United States. His company, Palantir, provides the surveillance and targeting software that intelligence agencies depend on. His fund made the largest single investment ($1 billion) in Anduril, the defence startup displacing legacy military contractors.

He provides the ideology that justifies the Great Houses’ rise, and the financial architecture that enables it.

The House of Bezos

Jeff Bezos, net worth approximately $290 billion, operates more quietly but no less structurally.

Amazon Web Services is the cloud backbone for the CIA, the US Department of Defence, and much of Western government computing. The state’s digital infrastructure runs on his servers. He owns the Washington Post, giving him direct influence over American political narrative. Blue Origin, his space company, receives $2 billion per year in personal funding.

Bezos demonstrates that you don’t need to be loud to be sovereign. You just need to own the infrastructure that sovereignty depends on.

The House of Saud

Mohammed bin Salman represents the Great House thesis in its purest form: a person who is the state while operating like a private investor.

He chairs the Saudi Public Investment Fund ($1.15 trillion in assets under management, targeting $2 trillion by 2030) and controls Aramco ($1.6 trillion market cap). The PIF owns Newcastle United, a majority stake in Lucid Motors, LIV Golf, and positions across Western equities, gaming, and technology.

The House of Saud is what happens when the Great House and the state become indistinguishable.

Why the Great Houses Are Winning

The economics of violence have turned. A $500 drone kills a $3 million tank. Non-state actors disrupt global trade with consumer-grade technology. The cost of projecting meaningful force has dropped from the scale of nation-states to the scale of well-funded private actors. The nation-state’s 400-year advantage, that only states could afford armies, is eroding faster than any government is willing to admit.

Private actors now control sovereign infrastructure. Starlink is the communications backbone of an active European war. AWS runs the CIA’s cloud. Palantir provides the analytical brain of Western intelligence services. Grok is integrated into the Pentagon’s classified networks. The state has outsourced its own nervous system to the Great Houses and has no backup plan if they defect.

Wealth concentration is at historical extremes. The top 0.001%, fewer than 60,000 people, control three times more wealth than the bottom half of humanity. Billionaires are 4,000 times more likely to hold political office than ordinary citizens. This is not influence over the state. This is direct occupation of the state.

The population is structurally incapable of resistance. Postman’s thesis, confirmed by every serious scholar of media over the past four decades, describes a population trained by algorithmic feeds to confuse entertainment with information, outrage with participation, and spectacle with politics. You cannot organise resistance when the medium through which you would organise is owned by the people you need to resist.

Technology favours the small and the fast. AI allows small teams to wield disproportionate analytical, creative, and operational power. A private AI laboratory can now outperform entire government agencies in specific domains. The bureaucratic, consensus-driven decision-making of democratic states cannot keep pace with private actors who move at the speed of a single person’s judgement.

The Case Against, and Why Most of It Doesn’t Hold

The strongest argument against this thesis is nuclear weapons. Nine states possess them. No private actor has come close. Nuclear deterrence remains the ultimate guarantor of state sovereignty, and no Great House can credibly threaten mutual assured destruction. The Great Houses will not replace the nuclear powers (the United States, Russia, China, the UK, France, India, Pakistan, Israel, and North Korea). They will operate alongside and within them, reshaping power from the inside rather than conquering from the outside.

The second argument is that the state can fight back. And the case of Xi Jinping and Jack Ma is instructive. When Ma’s power was perceived as threatening, Xi shut down Alibaba’s $37 billion IPO overnight, forced a corporate restructuring, and made the richest man in China disappear from public life for months. This is decisive evidence that a sufficiently motivated authoritarian state can crush its most powerful private citizen.

But this counterexample proves the thesis rather than refuting it. The countervailing force that exists in China has been effectively dismantled in the West. Western democracies spent forty years systematically transferring power to financiers, bankers, and entrepreneurs through deregulation, privatisation, tax arbitrage, and the legal fiction that corporations are persons with rights. The regulatory toolkit that once constrained robber barons (antitrust, nationalisation, punitive taxation) exists in theory but has atrophied in practice. The state is not fighting back. It is being absorbed.

I think technological determinism is, in this case, mostly sufficient. There will be exceptions. China is one. Perhaps Singapore, with its unusual state capacity, is another. But across the West and much of the rest of the world, the countervailing force has been hollowed out. The Great Houses are rising not because they overthrew the state but because the state gave its power away, piece by piece, over decades, and is now too weak to take it back.

Neither Utopia Nor Dystopia. Both.

The Sovereign Individual concludes that we are heading toward a libertarian utopia. A world of sovereign individuals freed from the coercion of the state, living wherever they choose, paying only for the services they use, answerable to no collective authority. Yanis Varoufakis, from the opposite end of the spectrum, argues in Technofeudalism that we are heading toward something far darker: a world where digital platforms extract rent from captive populations exactly as feudal lords extracted rent from serfs.

I think both are right. For different people.

For the Great Houses and their inner circles (the engineers, the investors, the operators) the future may indeed resemble a kind of libertarian utopia. There may be room for a managerial class. Jurisdictional arbitrage, golden visas, zero-tax residency in the UAE or Singapore, private healthcare, private education, private security, and private orbital communications. A world of exit and choice. The Sovereign Individual’s vision, realised.

For everyone else (those who cannot exit, who depend on public services, who live in the countries the Great Houses have moved their money out of) the future will feel like techno-feudalism. Rent extracted by platforms you cannot leave. Information curated by algorithms you cannot see. Power exercised by people you did not elect and cannot reach.

The medieval world had both. The lords lived in something approaching freedom and luxury while the serfs lived in something approaching bondage. The difference is that the serfs knew they were serfs.

In the new order, the serfs will be entertained.

What Matters Most: The Culture of the Great Houses

If this analysis is even partially correct, then the most important question in the world is not who wins the next election. It is who the Great Houses are. What they believe. What values they encode into the algorithms and platforms that increasingly mediate all human experience.

The dominant ideology of today’s Great Houses is a coherent bundle that scholars have labelled TESCREAL (Transhumanism, Extropianism, Singularitarianism, Cosmism, Rationalism, Effective Altruism, and Longtermism). Its tenets include:

Anti-democratic conviction. Thiel: “freedom and democracy are incompatible.” Curtis Yarvin (Mencius Moldbug), the intellectual architect of neoreaction, proposes replacing democracy with CEO-states, a model explicitly admired by influential Silicon Valley figures.

Technological messianism. A quasi-religious faith that technology will solve all problems, that acceleration is a moral imperative, and that those who urge caution are enemies of progress. The structure is theological: prophets (Kurzweil, Andreessen), saints (Hayek, Nietzsche), heresies (regulation, safety), and a salvation narrative (multi-planetary civilisation, the singularity, the defeat of death).

Longtermism. The belief that the trillions of potential future humans dwarf current populations in moral weight, which means that present-day suffering (poverty, inequality, exploitation) is morally secondary to ensuring the long-term survival of the species. This framework conveniently justifies any present-day concentration of power that serves the “long-term good” as defined by the people who already hold it.

Exit over voice. Seasteading, charter cities, cryptocurrency, golden visas, network states. A systematic preference for escape from democratic accountability rather than reform of it. If you don’t like the rules, leave. If you’re rich enough, build your own country.

Cognitive elitism. The explicit belief that a “cognitive elite” sees what others cannot, deserves to lead, and is held back by democratic masses too mediocre to understand what is good for them.

This is not a representative sample of humanity. It is a narrow ideological band, predominantly male, predominantly American, predominantly educated at a handful of elite universities, formally secular but structurally religious in their technological faith. And it is this culture, these values, these assumptions about human worth and the good life that are being encoded into the algorithms that determine what billions of people see, believe, and want.

When the Medici funded the Renaissance, their patronage shaped European art and philosophy for centuries. When the British Empire imposed English common law and Christianity, it restructured the cultures of a quarter of the planet. When American cultural hegemony spread through Hollywood, Coca-Cola, and the dollar standard after World War II, it defined what “modernity” meant for the entire world.

The Great Houses are doing the same thing. Faster, and through channels that are harder to see. They are not imposing culture through armies and missionaries. They are imposing it through algorithms and platforms. And they are doing it right now, in real time, while we scroll.

Why I Built The Great Houses

This is why I created The Great Houses: a project dedicated to tracking, profiling, and understanding the individuals and families who are accumulating sovereign-scale power.

The traditional lens for understanding the world is geopolitics. What are governments doing? What alliances are forming? Who is winning elections? This lens is increasingly inadequate. The decisions that will shape the next century are not being made in parliaments. They are being made in boardrooms in San Francisco, Riyadh, and Austin. Not by officials who must face voters, but by individuals who can exit any jurisdiction that displeases them.

If the Great Houses are disproportionately composed of one particular ideology, religion, culture, or worldview (and they are) then that worldview will over-represent in the future. Their values will become the default values of the infrastructure through which humanity experiences reality. Their assumptions about human nature, about what constitutes the good life, about who deserves power and who deserves to serve, will be encoded not in law, which can be changed, but in technology, which is far harder to reverse.

I think watching the Great Houses is more important than watching any government. Understanding what they believe is more important than understanding any party platform. Tracking their movements (their investments, their acquisitions, their alliances, their conflicts) is the most important geopolitical analysis of our time.

The nation-state is not dead yet. But it is being hollowed out from the inside, its functions transferred to private actors who are accountable to no one, while its populations are too entertained to notice.

The Great Houses are rising. And the world they build will reflect who they are.

This is the story of AI adoption right now. Hundreds of millions of people are using AI tools daily, not because those tools are delightful or even particularly accurate. They’re using them because the time equation is absurdly positive. And this clarifies what the most important metric of the AI era actually is: net time saved.

The formula is simple.

Net Time Saved = Time to do it yourself − Time to get AI output to your standard

That’s the metric. No complex product framework. No five-dimensional scorecard. Its the one number that tells you whether the tool earned its keep.

Here’s a real example. When I use Nume to run a monthly financial performance review of my business, it produces an analysis that would have taken me five hours to put together manually. I get the output, review it, make a few tweaks to tailor it for a specific stakeholder. Twenty minutes. My net time saved is four hours and forty minutes.

I’m getting hours of my day back. How could I not love that?

The real lesson: go big.

Net time saved has an important implication that most people miss. It rewards complexity.

If you use AI to fix a typo, you saved thirty seconds. If you use AI to draft a five-page financial report from scratch, you saved five hours minus twenty minutes of review. If you use AI to build a full scenario analysis modelling three different growth trajectories with sensitivity tables, you might have saved yourself two full days of work, even after two hours of corrections.

The more complex the task you hand to AI, the larger the net time saved, even with relatively small accuracy drops and review time increases. A simple task done at 95% accuracy saves you minutes. A complex task done at 70% accuracy saves you days.

This is the lesson that changes what you build next. Don’t build an AI solution for small improvements on the last iteration. Go for the hard problems. Go for the things that would have taken someone a full day, or that you wouldn’t have attempted at all. The time you get back is transformational.

For many complex tasks, the baseline isn’t perfect human output. The baseline is not doing the task at all. A founder who doesn’t have time to build a financial model isn’t choosing between a perfect model and an imperfect AI model. They’re choosing between an imperfect AI model and no model. An 80% accurate draft that exists beats a 100% accurate draft that doesn’t.

The “accuracy mirage” argument doesn’t hold up.

There is plenty of writing arguing that the productivity gains from AI are illusory because people spend too much time correcting errors. A widely cited Workday study found that employees spend 40% of their AI-saved time fixing mistakes.

That sounds damning until you do the actual math.

If AI saves you ten hours and you spend four hours fixing errors, you still saved six hours. Six hours you didn’t have yesterday. Six hours you can spend on work that actually requires your judgment, your relationships, your creativity. The critics frame the 40% as a failure. It’s not. It’s a 60% net gain on work that previously consumed your entire day.

The critics also conflate two very different kinds of errors. Correctable errors are things like formatting, phrasing, a number that’s slightly off. These cost minutes to fix. Structural errors are where AI gets the methodology wrong, or a mistake in one cell propagates through fifty rows of a financial model. These cost real time.

Both are solvable. At Nume, we’re building systems where correcting a structural error is as simple as fixing one number and watching it flow through every downstream calculation automatically. The same way a well-built spreadsheet model works. One correction, full propagation. That turns a structural error from a thirty-minute rebuild into a thirty-second fix, and it makes the net time saved equation even more lopsided in favour of AI.

People already get this.

Look at how people actually behave.

Every business hires junior employees who make mistakes. Nobody fires a junior analyst because their first draft needed revisions. You hired them because even with the supervision and correction, having them on the team saves you time. A junior analyst costs $60K a year and produces imperfect work for their first twelve months. Nobody questions the ROI because the net time saved across the team is obvious. AI does the same thing for $20 a month.

People have rapidly calibrated their expectations. Everyone who uses an AI tool in 2026 knows it makes mistakes. It says so in the fine print of every product. Users aren’t shocked by this. They’ve internalised it. They check the output, make corrections where needed, and move on. The same way they would with any capable but imperfect colleague.

The high-stakes exception.

This doesn’t mean accuracy is irrelevant. There’s a threshold below which time saved turns negative. And in high-stakes domains, the penalty for errors extends beyond correction time.

A wrong number in a board deck doesn’t just cost twenty minutes to fix. It costs credibility. An incorrect medical dosage costs lives. In these contexts, net time saved needs a risk adjustment:

Risk-Adjusted Net Time Saved = Time saved − Correction time − (Probability of error × Cost of that error)

If a consequential error has a 5% chance of slipping through review and the downstream cost is ten hours of damage control, that’s thirty minutes of expected penalty on every task. For most complex tasks, the time saved still dominates. But the formula forces you to be honest about it.

Even in high-stakes domains, though, the comparison isn’t AI versus perfection. It’s AI plus human review versus human alone. Humans also make errors. The question is whether the combined system produces better outcomes in the same time budget. In most cases, it does.

Why this metric wins in the AI era

Net time saved captures things other metrics miss.

It’s the only metric users can feel. Nobody feels a 2% accuracy improvement on a benchmark. Everyone feels getting their evening back. Visceral value drives word of mouth, retention, and willingness to pay.

It compounds. Five hours saved per week is 260 hours per year. That’s six and a half working weeks. For a solo founder, that’s the difference between burning out and building a company.

It’s the great equaliser. A solo founder with AI that saves them twenty hours a week has the operational capacity of a two-person team. The real disruption isn’t better software. It’s more time for people who have the least of it.

It predicts willingness to pay. People will pay almost anything for a tool that reliably saves them hours per week. Price sensitivity drops dramatically when the value is measured in reclaimed time.

It aligns builder and user. When your north star is net time saved, every product decision gets clearer. Should you improve accuracy or speed? Whichever saves the user more time. Should you add a feature or polish an existing one? Whichever saves the user more time.

Measuring it across your product.

For product teams who want to track this seriously, here’s the portfolio-level formula:

Total Net Time Saved = Σ (Time saved per task − Correction time per task) × Frequency of that task

This captures something important: saving ten minutes on a task someone does daily (fifty minutes per week) is more valuable than saving two hours on a task they do monthly (thirty minutes per week equivalent). Frequency matters as much as magnitude.

Ask your users two questions. How long would this have taken you without us? How long did you spend correcting our output? That gap is your value. Everything else is a vanity metric.

The promise of AI is not delight. It’s time saved.

The B2B products that win the AI era will not be the ones with the highest accuracy on benchmarks. They will not be the most delightful or the most polished. They will be the ones that maximise net time saved, consistently, across the tasks that matter most to their users.

The ultimate benefit of AI is not that it will produce the most entertaining experience. It’s not that it will generate the most accurate financial reports, though it might. The ultimate benefit is that it gives you your life back.

Go big. Tackle the complex problems. Let the cleanup cost you minutes so the output saves you days.

That’s the metric. Everything else is a trailing indicator.

The ongoing humiliation

Every Chinese child is taught about the century of humiliation. It’s the roughly one century between the burning of the forbidden city, through the great leap forward, up until the fruits of Deng’s reforms were realised. For muslim-majority countries, the humiliation began long before the Chinese forbidden city was burnt to ashes and it continues unabated into the present. Most muslims live in, to borrow a phrase from Trump, “shit hole” countries that most of them are trying to escape. My father, like many Pakistanis, left that country for a better life abroad and I’ve benefited immensely from that decision. However, today we see many immigrants escape their “shit hole” countries only to attempt the resurrection of their home culture in the new place. This resurrection of an idealised past is done under the protection of individual rights in countries like the USA or UK, though many of these immigrants seldom believing in such rights themselves.

Nobody is waiting around for scientific or academic innovations from the Muslim world, those who do go on to achieve great things do so by going to the West where they enjoy freedom of expression.

Why hasn’t the muslim world reformed? China certainly reformed. It made its culture accommodate capitalism and then the internet. Despite being nominally communist, the Chinese were able to adapt to the realities of capitalism and the internet and have even come to transcend the West in these domains. It might be called the Chinese Communist Party but they are better at Capitalism than most Europeans who get relatively poorer every year. If Muslim cultures are to have any hope in the modern world, they must make similar cultural changes. They must reform to accommodate the reality of the internet and capitalism. Any culture that is not compatible with capitalism and the internet is doomed to poverty.

Capitalism and the internet are giant assimilation machines into the global culture.

You can’t opt out of capitalism or the internet. It’s everywhere. As Marc Andreessen noted, “the internet is a giant assimilation machine” into the global culture, mainly Western culture - “it will rip every kid out of their own culture”. Capitalism is much the same, it’s like a giant assimilation machine into a globalised culture that fosters free trade.

What will muslims contribute to the internet’s culture?

Muslims largely hail from countries where women are humiliated by men. The men are humiliated by their own governments. The state is humiliated by the West. Draw a straight line from Casablanca to Islamabad on a map and there is not a single country in between that respects the dignity of a human being, the freedom of the individual or provides economic opportunity to the masses. Not a single one.

Young muslim men in England combine the worst traits of the British underclass, the chavs, with the puritanical elements of Islam. The worst of both worlds. They live in free societies but support policies that would turn their new homes into the places that their parents escaped. When Western commentators attack Muslim young men for their failure to assimilate, the young men appeal to values of human rights and individualism, values that they themselves do not believe in nor do they live by. No wonder, that nobody takes them seriously. How could you take such people seriously?

Almost everything is downstream of the culture.

Intolerance and dogmatic thinking are the norm in the muslim world. Reform and contrarian thinking rare, if not outright banned.

The repressive culture that forces the separation of men and women leads to a fanaticism in young men who are taught that “their” women are off-limits but the foreign jezebels are fair game. This manifest itself in incel culture and the harassment of women. Taken to extremes, it manifests itself in the grooming gangs of northern England.

Pakistan, Syria, Dafür, Myanmar, Gaza, Afghanistan, Iraq, Yemen.. the list goes on. Most of the violence in the contemporary world can be attributes to muslims of one sub-culture killing Muslims of another one, albeit often with international backers with their own agendas, see Syria and Libya. In a minority of cases, it is simply more powerful cultures engaged in power plays, see India and Israel. I am sympathetic to the arguments that the meddling of the United States is the driver of many problems in the muslim world but it can’t explain everything. It is too tempting to resort to victimhood narratives

Even in the rich Gulf Arab states where a group of muslim men have dictated the terms of the global oil market, literally the most important commodity to civilisation, there are scant achievements to be proud of. The Gulf Arabs have created shoddy replicas of Western cities. By emulating the buildings and material wealth of the West, they hoped to gain membership to that club. But the underlying culture has not changed massively. There are no world class universities, there is no export of culture to the rest of the world. The Gulf Arabs are (hopefully) slowly learning that the West has not achieved its advantages because they figured out how to build skyscrapers, they are ahead because they have a culture that values advances in science, technology and arts.

Brutalised, degraded, beaten. Men and women from Muslim majority countries suffer humiliation all the time. Abu grain twenty years ago. Israeli internment camps today. But that’s just the tip of the iceberg. The humiliation from westerners pales in comparison to what the illiterate generals and mullahs who run these countries do to their own people

What is not the root cause of the problems

It is tempting to blame racism since, many of us, wear the face “the other”. We can be easily identified in a crowd. But that has not held back South Indians who have embraced capitalism and excelled among the managerial classes in the West. It has not held back Chinese. Both communities earn salaries in the US that are almost double the average white American.

It is tempting to blame colonialism but South Korea and Ireland are now richer than the nations that colonised them. India is on the hockey stick curve when it comes to GDP per person.

It is tempting to romanticise the old-world culture, to say that, yes we may have problems but we have superior family values that transcend those of the West. That we have superior moral values and we should simply stick to our guns and eventually, we’ll become rich. As a religious Muslim at a wedding recently told me, “our problem is that we don’t have enough belief in Islam”. Many Muslims will put a spin on the old George Dubya Bush line that “they hate is because of our freedom”, which instead becomes “they hate us because we are the righteous ones”. This is the LARP that so many western leftist Muslims love to engage in.

These are fantasies, pure and simple. The root causes go much deeper. There is no nobility in staying impoverished.

So where does the blame lie?

Let’s start at the top, It is tempting also to put all the blame on the weak and dogmatic leadership in the East, and here I do have some sympathy for the arguments. Muslim countries are stuck with some leaders who are wordcel fanatics in the mold of Khamenei who, in a desperate attempt to win an asymmetric war, condoned suicide bombing which lead to a wave of horrific death and destruction across the Muslim world, which continues today. There are other leaders who come in the mold of illiterate rent-seeking strongmen who are their to keep the backward masses in check and earn rental payments. Think Sisi in Egypt or anyone who has run Pakistan. Then there are the Islamic idealists like Gaddafi, or on the moderate end, Erdogan. They attempt to make themselves the defacto leaders of the “ummah” or Islamic world and pursue Islamist policies to some extent. Leadership is a big problem, but it’s not the root cause.

It is tempting still to use a variation of the aforementioned George W Bush “They hate us because our freedom” argument. This is often heard from leftists who claim that Islam being the greatest religion is of course a target of those who are not as righteous. This is the “They hate us because we are the righteous ones”. Of course, we’re targets, it’s because we’re right about everything. This has to be the dumbest argument of the bunch. I have absolutely no time for this nonsense.

Political Islam as Communism

To get at the root problem, it is helpful to compare political Islam to communism

The great Czech leader Vaclav Havel asked his readers in his essays why it is that a butcher feels the need to display a pro communism banner in his shop front. Is he a genuine supporter of the cause? Was he inspired by reading the German Ideology by Marx and Engels? Not at all. It is the difference between, what Timur Kuran calls private truths and public lies. In public the butcher is a communist, a card carrying member. But in private he may express different opinions. In public, he believes the Soviet Union is the greatest country on earth. In private, he wants his children to move to the United States. In public the young man in Czechia might wear something that meets the approval of his elders, in private he wants the blue jeans from America. Something similar can be said about political Islam. Does the shop owner in Karachi really believe in transforming Pakistan into a medieval Islamic emirate? Or does he just say this publicly and believe something else in private?

Both political Islam and communism require people to hide their true beliefs, it requires preference falsification, the idea that people won’t share their true preference in public.

The Root Cause

Everything is downstream of culture. There are two elements of modern living that are here to stay and that the culture must accommodate, that is capitalism and the internet. Any culture that cannot live alongside capitalism and the internet is doomed to failure

The culture suffers from four key defects that make it incompatible with capitalism and the internet; no respect for individual expression, the oppression of minorities, the oppression of women and no culture of excellence. To make progress, Muslim countries must create a culture that respects individual freedom, respects the rights of women and minorities and creates a culture of excellence.

Tradition imprisons us all. Tradition is a prison that prevents us from being a part of the mainstream international culture. We must do away with many traditions or adapt them to capitalism and the internet. The traditional culture is the problem.

How do we get out?

As Monmohan Singh addressed the Indian parliament in 1991, he looked on at the growth in Taiwan, South Korea and China and quoted victor Hugo “no power on earth can resist an idea whose time has come”. The time for cultural reform is upon us and we must seize it. We must demand that leaders make the necessary reforms. As individuals, we should shame others who do not live up to the ideals of individual expression and excellence. We cannot wait, there is no time like the present.

For the past seventy years, humanity has been converging. Same brands. Same malls. Same Netflix queue. A street in Riyadh looks like a street in Raleigh looks like a street in Rotterdam. Anglos built a global cosmopolitan monoculture around consumerism and exported it to every corner of the planet.

It worked. Spectacularly.

And now it’s falling apart.

Everywhere Is Nowhere

The global monoculture succeeded beyond anyone’s imagination. McDonald’s operates in 100+ countries and feeds 1% of the world every day. Instagram has 2 billion monthly users. English is the lingua franca of business, science, and entertainment. Kids in Bangladesh who’ve never set foot in a school speak English in an American accent because they’ve consumed so much mono-culture youtube. For the first time in human history, a teenager in Lagos and a teenager in London consume substantially the same cultural diet.

But when everywhere looks the same, nowhere feels like home.

Frank Wright, a British reporter, captured this in a street interview that went viral in May 2026, 40 million people watched it. Wright described England as having become “a borderless international supermarket populated by strangers who are hostile to one another.” He said he’d “rather live in a nation and not a market.”

40 million views. A nerve was struck.

The monoculture flattened everything. It gave us efficiency, optionality, and consumer choice. What it took in return was belonging. The mall replaced the town square. The algorithm replaced the conversation. The global citizen replaced the neighbour.

People tolerated this trade-off for decades. I think the tolerance is ending.

The Splintering Is Already Happening

Here’s what the data actually shows.

32 million Discord servers exist today. 2.8 million subreddits. 70 million active Facebook Groups. The average user belongs to 8.5 different groups. 115 million YouTube channels. 3.5 million podcasts. 200+ streaming services.

These are identity structures. A 2024 Pew Research study found that 68% of internet users now prefer platforms fostering “deep, meaningful interactions” over broad social networks. People are actively choosing small and meaningful over big and meaningless. Niche over mainstream. Tribe over nation.

Here’s the weird thing that’s hard to articulate but instantly recognisable if you’ve experienced it: meeting someone who’s a deep, devoted member of a completely different online community, like a different subreddit, a different corner of the internet, is like meeting someone from a different country. Maybe more so. Because most nationalities have blurred into the same global monoculture. The person across the table from you at dinner shares your passport but inhabits a completely different reality. Meanwhile, someone 5,000 miles away who found the same obscure corner of the internet as you? They get you in ways your neighbours never will.

That’s the future.

Why Now? The Collapse of the Old Identity Structures

People don’t stop needing belonging. They just find it somewhere else.

The old structures are failing:

National identity. Only 41% of Gen Z feel proud to be American, versus 83% of the Silent Generation. Each successive generation identifies less with the nation they were born into.

Institutional trust. At historic lows across every institution measured. Government, media, church, business. The Edelman Trust Barometer shows a 30-point gap between high- and low-grievance populations globally. Gen Z trusts no institution by majority except science.

Monoculture. In 1983, the M*A*S*H finale drew 106 million viewers out of a population of 234 million. Nearly half the country watching one show. There were 3 broadcast networks. Today, broadcast TV commands less than 20% of viewership. No single piece of content commands mass attention.

The old containers of identity are draining. Nation, shared culture, institutional trust. The void they leave behind is enormous.

And the internet is filling it.

The Tribes Are Already Here

This isn’t theoretical. Let me show you what’s forming right now.

Bryan Johnson and the Don’t Die Tribe

Bryan Johnson sold Braintree Venmo to PayPal for $800 million. Then he spent millions per year trying to reverse his biological age and documented everything publicly.

The result is a full-blown tribe.

6 million followers across platforms. A Netflix documentary (Don’t Die: The Man Who Wants to Live Forever). A Don’t Die app with global meetups. A Don’t Die Summit that sold out 1,500 tickets. The Rejuvenation Olympics, a competitive leaderboard with 4,000 participants measuring who can reverse aging fastest. Blueprint raised $60 million from Kim Kardashian and the Winklevoss twins. His top-tier “Immortals” program charges $1 million per year for 3 clients.

Johnson is a tribal leader with a belief system. “Don’t Die” is a worldview. It has rituals (the protocol), a community (the app and summits), competition (the Rejuvenation Olympics), identity markers (the biomarkers), and a moral framework (longevity as civilisational purpose).

Praxis: The City-Building Tribe

Dryden Brown is 28 years old and raised $525 million to build a new city.

Praxis started as an online community, a Discord essentially. It now claims 151,068 “citizens” from 80 countries. Companies founded by Praxis members have an aggregate valuation of $1.1 trillion. 12,000 people are on a waiting list to physically relocate.

The location: Atlas, about an hour north of Santa Barbara, on Vandenberg Space Force Base. A “defence-focused spaceport city.”

Zero physical residents so far. The gap between digital citizens and actual bricks is enormous. But $525 million in committed capital is not a crazy fantasy. Coinbase’s venture arm invested in January 2025. Brian Armstrong is a believer.

Praxis is testing the question: can an online tribe become a physical city? I’m bullish.

Balaji Srinivasan: The Network State

The Network State (2022) argues that nations will increasingly be founded online first, then acquire physical territory. The progression: Network Union (online community) → Network Archipelago (scattered physical nodes) → Network State (diplomatically recognised entity).

His Network School, a residential community on an island near Singapore, started with 150 people in 2024 and expanded to 400 by mid-2025. He’s planning campuses in Miami, Dubai, and Tokyo.

Government officials from Singapore, Dubai, Abu Dhabi, El Salvador, and Palau are engaging with the concept.

These are not theoretical conversations anymore.

The Special Economic Zone Explosion

In 1975, there were 79 special economic zones across 25 countries. Today there are over 7,000 across 145+ countries. That’s an 88x increase in 50 years.

Saudi Arabia launched 4 new SEZs in 2023 offering 0% corporate tax for 50 years. The UAE has 45 free zones. 47 of 54 African countries have implemented SEZ programmes. s

These zones matter for the tribalization thesis because they provide the legal substrate for experimentation. Every new SEZ is a new set of rules. Every new set of rules is a potential attractor for a community that wants to live differently.

The network state movement is essentially arguing: take the SEZ model further. From economic autonomy to governance autonomy. From a zone to a city. From a city to a nation.

Remigration: The Involuntary Catalyst

There’s another force that could accelerate tribal formation, and it’s less comfortable to talk about.

Remigration, calling for the mass return of immigrants and their descendants to countries of origin, has moved from fringe to mainstream European politics in less than three years.

Austria’s chancellor launched “Operation Fox.” Germany’s AfD held a secret meeting in Potsdam to discuss remigration plans, exposed by journalists in January 2024. Geert Wilders in the Netherlands called for “large-scale remigration programmes” in parliament. The European Parliament approved legislation enabling offshore detention centres and two-year detention periods. The Save Europe Act, a citizens’ initiative demanding a total halt to non-European immigration, gathered 100,000 signatures in three days.

The US State Department created an “Office of Remigration” in May 2025.

Whether or not these policies are implemented at scale, the political direction is clear. And if even a fraction materialise, you’re talking about hundreds of thousands — potentially millions — of people, many second- or third-generation, who don’t really fit into their countries of origin any more either. They’re too European for where they came from and, apparently, too foreign for where they are.

Where do they go?

This is exactly the kind of dislocation that creates demand for new communities. New cities. New economic zones. New tribes. People who find themselves between worlds are precisely the people most likely to build new ones.

The heartlands those people are originally from are full of underdeveloped land and governments eager for investment. The ingredients for new charter cities, new planned communities, new economic zones are all there. The people are there. The capital is there. The legal templates exist.

This isn’t speculation for the distant future. The political machinery is already in motion.

How AI Is Accelerating Everything

I don’t think AI’s primary contribution to tribal fragmentation is leisure time. Every past automation wave increased total work rather than reducing it. I see no reason AI will be different on this front.

But AI is doing something far more powerful than freeing up time. It’s building the entire infrastructure of tribal life.

AI as Tribal Matchmaker

TikTok’s recommendation algorithm doesn’t just find content for you. It sorts you into a tribe.

The platform has generated hundreds of “-core” aesthetics that function as micro-identities: cottagecore, dark academia, goblincore, blokecore, clean girl, coquette. Each one started as a content tag and became an identity that people adopt, signal, and defend.

No two For You Pages are alike. The algorithm creates individualised reality tunnels that route people into specific communities. Research from the University of Michigan confirms that “TikTok’s unique format facilitates the rise and proliferation of diverse subcultural communities.” YouTube removed its Trending page entirely in July 2025, shifting to personalised recommendations to capture micro-trends across different fandoms.

The algorithm is the most powerful tribal matchmaker in human history. And it’s getting better every day.

AI Makes One-Person Tribal Economies Viable

The share of new startups with a solo founder jumped from 23.7% in 2019 to 36.3% in the first half of 2025. Over 48,000 solo-founded startups launched in 2025, up 140% from 2024. As of early 2026, 38% of seven-figure businesses are led by solopreneurs who replaced traditional hires with AI workflows.

Pieter Levels runs NomadList, RemoteOK, and PhotoAI generating $3-5M annually with zero employees. Danny Postma built HeadshotPro to $300K/month working solo. Dario Amodei, CEO of Anthropic, gives a 70-80% probability that a one-person billion-dollar company will happen in 2026.

When the minimum viable business can serve 100 customers instead of 10,000, you can build a business that serves exactly one tribe. A solo founder can build a product for vegan CrossFit coaches or fractional CFOs serving law firms and make it work. Each niche business reinforces the identity of its community. The product becomes tribal infrastructure.

The Tribal Taxonomy

Not all tribes are alike. Here’s what’s forming:

Creator Tribes. A leader with concentric rings: inner circle of superfans, paying members, casual followers. Economics: direct patronage via Patreon, Substack, Skool. The creator economy is $252 billion and growing at 23% annually. 207 million+ active creators worldwide. These are tribal leaders, whether they know it or not.

Mission Tribes. Organised around a shared cause rather than a person. Bryan Johnson’s Don’t Die movement. Effective Altruism. The longevity community (Vitalia, Infinita). The solarpunk movement. These tribes have the strongest binding force because the mission provides existential purpose. This is the flavour of tribe I’m most bullish on.

Professional Guilds. Organised around a craft, with mentorship hierarchies and mutual economic support. Indie Hackers. On Deck. YC alumni. CrossFit affiliate owners. The medieval guild model, reborn digitally.

Lifestyle Tribes. Organised around how you live. Digital nomads (WiFi Tribe). Van-lifers. Minimalists. Biohackers. These are the ones most likely to create physical spaces — co-living, intentional communities, popup cities.

Governance Tribes. Explicitly seeking sovereignty. Praxis. Próspera. The Network State movement. Liberland. The most ambitious category and the furthest from realisation.

Fan Tribes. Organised around shared cultural consumption. K-pop fandoms. Twitch communities. Gaming guilds. The largest tribes by raw numbers, and the ones most successfully making the leap from online to offline (VidCon, TwitchCon, comic conventions).

Diaspora Tribes. People caught between worlds, too European for where they came from, too foreign for where they are. This category is about to grow significantly if remigration policies advance.

The Business of Tribes

The emerging opportunities are:

Tribal governance tools. If thousands of communities need to make collective decisions, the tooling is primitive. DAOs proved the demand ($26-28B in treasuries) but the UX is terrible. Whoever builds simple governance for non-crypto communities wins big.

Physical space development for tribes. Co-living is $4B heading toward $30B+. But most co-living is generic — affordable housing with a communal kitchen. Purpose-built tribal spaces (a building for longevity enthusiasts, a neighbourhood for indie game developers, a co-living for climate tech founders) barely exist. The demand is clearly there. The supply isn’t.

Community-led growth infrastructure. Companies using community-led growth see 32% lower customer acquisition costs and 46% higher customer lifetime value. 24% of companies can now quantify the financial value of their community. Of those, nearly half report $1M+ in impact. But the tooling to build, measure, and scale community-led growth is fragmented.

Inter-tribal services. As tribes proliferate, they need ways to negotiate, trade, and coordinate with each other. Translation layers between tribal economies. Reputation systems that travel across communities. Dispute resolution. This market doesn’t really exist yet.

Where This Is Actually Heading

The fragmentation is real, it’s measurable, and it’s accelerating. But the timeline matters, and the shape is different from what the most enthusiastic advocates imagine.

The digital fragmentation is essentially complete. By 2035, most people under 40 in developed countries will derive more of their identity from chosen online communities than from nationality, religion, or employer. This is a straight-line extrapolation of current trends. I’d bet on it heavily.

Physical tribal spaces will grow but remain niche for the next decade. Co-living, intentional communities, and popup cities will grow from thousands to tens of thousands of participants. Not millions. Not yet. Physical space has too much friction.

One or two charter city projects will reach 10,000+ residents within 15 years. Próspera, Praxis, or something we haven’t heard of yet. The legal infrastructure (7,000+ SEZs), the capital ($525M+ for Praxis alone), and the demand are all there. But diplomatic recognition of anything resembling a network state is a 2050+ event at the earliest.

AI won’t give us leisure. It’ll give us independence. Rather than creating free time, AI will make it viable for more people to earn a living serving a niche community. The minimum viable business drops from 10,000 customers to 100. The minimum viable tribe drops from requiring a team to requiring one person with the right tools. More tribes become economically sustainable. The flywheel accelerates.

The biggest risk isn’t that this doesn’t happen. It’s that it happens unevenly. A world of self-selected tribes can easily become a world of echo chambers, gated communities, and jurisdictional arbitrage for the wealthy. Tribes for the rich. Atomisation for everyone else. The infrastructure to prevent this doesn’t exist yet.

Remigration, if it materialises at scale, will be an accelerant. Millions of people in search of belonging, with capital, skills, and no obvious home. Exactly the conditions that produce new communities, new cities, and new economic zones. The heartlands those people came from are full of underdeveloped land and governments eager for investment.

The Bottom Line

The global monoculture reached its peak. It made everywhere the same. And because everywhere is the same, people are searching for somewhere that feels like theirs.

The internet gave them the tools to find each other. AI is giving them the tools to build together, to manage communities, create culture, run micro-economies, communicate across languages, and construct shared realities.

The tribes are coming. Millions of them. Most will be digital. Some will become physical. A handful will attempt sovereignty.

The nation-states aren’t going anywhere. But they’re going to share the stage with something new; fluid, voluntary, identity-rich communities that people choose rather than inherit. Communities where you belong because you want to, not because you were born there.

For seventy years, the world converged. Now it’s diverging again. Not back to the old tribes of blood and soil. Forward, into something we haven’t seen before.

The great tribal fragmentation of humanity has begun.