CHAPTER 4 — WINNERS AND LOSERS (THE SI’S GEOPOLITICAL PLAYBOOK)
In the first chapters we sketched the fork in the human story: people who use AI as a feature, and people who use AI as an operating system. This chapter is for the second group. It answers the “Winners and Losers” brief in your living-book prompt set , but it does it from the perspective of the AI-enabled Sovereign Individual (SI): someone who embraces the agentic life, runs a personal stack of models and tools, and treats geography as something you can choose rather than something you inherit. That single shift—from “I live where my employer is” to “I live where my system performs best”—turns the map upside down.
The old map was about jobs, wages, and the accident of birth. The new map is about governance, energy, and optionality. It is about which places can compound your agency rather than tax it away in friction. And it’s about which places will be forced into reactive politics by the stressors AI amplifies: migration, inequality, and legitimacy crises. The “winners” in this chapter are not morally superior. They are jurisdictions (and a few regions) whose institutions, incentives, and infrastructure are positioned to attract talent and capital, scale compute and automation, and keep social order intact while the global labor market is being rewritten. The “losers” are the places where the opposite dynamics stack: low state capacity, high demographic and climate pressure, weak capital formation, and growing exposure to displacement and instability.
Before the lists, you need the SI lens. For an SI, a country is not just a flag and a culture. It’s a bundle of APIs you will call every day: the residency system, the banking system, the telecom system, the legal system, the tax system, the security system, and—more than ever—the energy system. Each one can either be a multiplier (low-friction, predictable, digital-first) or a throttling valve (slow, opaque, arbitrary, corrupt, paper-based, politicized). Your “life uptime” is the product of how many of these APIs work reliably, and how catastrophic it is when one fails.
What changed since the early “LLM era” and the launch of agentic platforms like OpenClaw is that the state is beginning to regulate not just AI “as software,” but agentic AI as delegated action in the real world. Singapore’s January 2026 launch of a Model AI Governance Framework specifically for Agentic AI is an early signal that governments are internalizing the new risk surface: agents that can take actions, touch data, and make changes, not just generate text. (Infocomm Media Development Authority) That matters for SIs because it foreshadows a bigger pattern: the best-run places won’t only be “pro-AI” or “anti-AI.” They’ll be pro-accountability, pro-testing, and pro-trust, because that’s what lets agentic systems expand into finance, health, education, and government without imploding public confidence.
Now add the hidden variable: migration pressure. UNHCR reported 123.2 million forcibly displaced people worldwide at the end of 2024. (UNHCR) Mid-year 2025 reporting still put forced displacement above 117 million. (UNHCR) Those numbers are not abstract. They are an upstream indicator of political volatility in destination regions and border states, and they shape the “social cohesion tax” that even rich countries pay in the form of polarization, security spending, and institutional strain. As AI and robotics expand, the risk is not simply “job losses” in wealthy countries; it’s the widening gap between regions that can convert compute into productivity and regions that can’t, which can intensify displacement and secondary migration. Reuters has reported forecasts of additional displacement driven by conflict and instability, and those dynamics compound when aid budgets tighten and wars linger. (Reuters)
So here is the playbook: if you are building an SI life, you are choosing not only a lifestyle, but a stress regime. You’re choosing which governments are likely to remain functional under stress, which societies can stay cohesive as inequality spikes, and which places can keep the lights on—literally—while data centers, AI training, and robotics raise the floor for “baseline civilization energy demand.” This is why the lists tilt so heavily toward jurisdictions that combine (1) rule of law, (2) serious AI/compute investment, (3) reliable energy and infrastructure, and (4) a demonstrated willingness to compete for talent and capital.
THE TOP 10 WINNERS (COUNTRIES AND REGIONS) FOR THE NEXT 5–10 YEARS
Winner #1: The United States (especially the “compute + energy” corridors, not a single city). The U.S. is the gravitational center of frontier AI capital formation right now, and that matters because frontier AI is not evenly distributed; it clusters where investment, talent, compute, and procurement interact. Stanford’s 2025 AI Index reported U.S. private AI investment reached $109.1 billion in 2024—nearly 12 times China’s $9.3 billion and far ahead of other regions. (Stanford HAI) That investment dominance is not just “Silicon Valley vibes”; it translates into faster tooling, deeper labor markets for high-agency operators, and a denser network of counterparties who can pay for the kind of leverage SIs build. Layer on industrial policy: the CHIPS and Science Act created a subsidy and incentive structure aimed at pulling semiconductor manufacturing and advanced packaging back onshore, and the U.S. Department of Commerce has been issuing major CHIPS awards to firms like Samsung. (U.S. Department of Commerce) Even with political volatility and renegotiations of grants under a new administration reported by Reuters, the strategic direction—more domestic compute capacity, more domestic chip supply resilience—remains a bipartisan national security priority. (Reuters) For SIs, the U.S. is the “high ceiling” jurisdiction: you can build faster, sell faster, and access deeper capital markets than almost anywhere. The tradeoff is that it is also the “high variance” jurisdiction: tax complexity, healthcare costs, and political swings can create sharp regime changes. If you are an SI who wants to live in the U.S., you treat it like a powerful but noisy operating environment: you optimize for redundancy (legal structure, insurance, geographic diversification inside the country) and you accept that the U.S. is less a “stable sanctuary” and more a “high-throughput engine room.”
Winner #2: The United Arab Emirates (Dubai + Abu Dhabi as a purpose-built talent and capital magnet). The UAE is a prototype of the new competition: a state that treats talent, capital, and infrastructure as things you can deliberately import and assemble. From the SI perspective, the headline feature is still true as of late 2025: the UAE does not levy personal income tax on individuals. (U.AE) The deeper story is that the UAE is building the institutional scaffolding of an AI hub: residency pathways (including the Golden Visa system) are formalized and publicly documented, and they are designed to attract investors, skilled professionals, and founders. (U.AE) At the same time, Abu Dhabi has been pushing a compute-and-capital strategy at a national scale. In 2024, the AIATC announced MGX, with Mubadala and G42 as foundational partners, and described an investment focus on AI infrastructure, semiconductors, and core AI technologies. (g42.ai) The Middle East Institute described MGX as a $100 billion AI infrastructure investment firm launched by Mubadala and G42, positioning the UAE among the small set of countries thinking in “AI infrastructure at scale.” (Middle East Institute) State media has also highlighted the scale of domestic AI-related investments and partnerships with global firms, though you should treat any single-number claims as directional rather than audited truth. (WAM) The tradeoff for SIs is that the UAE is a high-control environment with a distinct legal culture; it is extremely business-friendly in some dimensions and extremely non-negotiable in others. But if you define “winner” as “place that can still attract mobile, high-leverage people in an AI-automated world,” the UAE is near the top of the list because it is explicitly optimized for that outcome.
Winner #3: Singapore (the high-trust, high-competence “agentic governance” hub). Singapore is what happens when a state treats competence as a national resource. It is not the cheapest place. It is not the largest market. But it is one of the best “reliability environments” for operating an SI life: low corruption, predictable enforcement, strong financial infrastructure, and an instinct to build frameworks rather than improvise. Singapore’s National AI Strategy page was updated in January 2026, and the country has been explicit that AI is a strategic priority rather than a side project. (Smart Nation Singapore) The most relevant signal for the agentic era is that Singapore’s IMDA announced a new Model AI Governance Framework for Agentic AI at Davos in January 2026, framing it as guidance for responsible deployment of AI agents. (Infocomm Media Development Authority) This matters because it’s a preview of how “serious” jurisdictions will compete: not by banning agents, but by making them deployable at scale with trust, testing, and accountability. For SIs, Singapore is a place where you can build systems that touch regulated domains (finance, health, education) with less existential fear that policy will become arbitrary. Taxes are not zero, but they are legible; IRAS notes the highest personal income tax rate is 24% (as of a December 2025 update). (Default) The tradeoff is cost of living and climate. But as a “headquarters for an agentic life,” Singapore is one of the most structurally solid bets in the next decade.
Winner #4: Switzerland (the “rule-of-law fortress” that stays boring on purpose). In an era of acceleration, boring is an asset. Switzerland is consistently ranked at the top of global innovation and competitiveness metrics, and while indices aren’t destiny, they often correlate with the traits SIs actually need: infrastructure quality, institutional stability, and predictable enforcement. WIPO’s Global Innovation Index 2025 places Switzerland at #1. (WIPO) IMD’s 2025 World Competitiveness Ranking also puts Switzerland at the top. (IMD Business School) For high-net-worth SIs, Switzerland still offers unique tax regimes in specific conditions; Switzerland’s federal finance administration describes expenditure-based (lump-sum) taxation for certain foreign nationals who are domiciled in Switzerland but not gainfully employed there. (efd.admin.ch) More importantly, Switzerland continues to signal policy predictability; for example, Swiss voters rejected a proposal for a steep inheritance tax on the super-rich in late 2025, reinforcing the perception of stability in the fiscal environment. (Reuters) The tradeoffs are cost, social integration, and immigration friction. But as a “safe anchor node” in an SI’s multi-jurisdiction strategy, Switzerland remains a top-tier winner because it is optimized for continuity rather than headlines.
Winner #5: The North Sea / Northern Europe Competence Belt (Denmark + Netherlands + Sweden + neighbors as a regional cluster). If you want to see what “state capacity” looks like in practice, look at the countries that can build infrastructure, maintain social trust, and run digital government without drama. Denmark and the Netherlands in particular show up repeatedly in competitiveness and governance discussions. IMD’s 2025 competitiveness ranking places Denmark in the global top tier, and the Netherlands remains a consistent high performer. (IMD Business School) The reason this region is a “winner” is not that it has the world’s lowest taxes (it doesn’t); it’s that it tends to have high functional quality and durable institutions—qualities that become more valuable as the world becomes more volatile. From an SI perspective, it’s also a region that benefits from the EU’s industrial reorientation: Europe is explicitly trying to increase semiconductor resilience via the European Chips Act, which entered into force in September 2023 and sets a target to reach 20% global market share by 2030. (Digital Strategy EU) The tradeoff is that Europe is also building a stricter regulatory environment for AI; the EU’s AI Act is part of that framework and has been updated on the EU’s own digital strategy pages as recently as January 2026. (Digital Strategy EU) That regulation can be an innovation drag in some contexts, but it can also be a trust accelerant if you’re deploying agentic systems in sensitive areas. Practically, the Netherlands remains a strong “legal + logistics” node for skilled talent and companies; the Dutch immigration authority outlines the highly skilled migrant residence permit pathway (with the usual constraints around recognized sponsors). (IND) If you can tolerate higher tax and compliance, this region is a winner because it stays functional under stress.
Winner #6: Ireland (the English-speaking EU bridge with a corporate and talent flywheel). Ireland is a classic example of a small country leveraging policy to become a global node. It has concentrated exposure to multinational flows, which is both a strength and a risk, but it remains structurally attractive because it combines EU access, an English-speaking environment, and a long-standing pro-business posture. Ireland’s Department of Finance has stated that Ireland continues to apply a 12.5% corporation tax rate for businesses outside the scope of the global minimum effective tax rate rules (with scope defined by the €750m threshold), while implementing the 15% effective rate for in-scope groups. (gov.ie) Ireland’s Revenue describes the OECD Pillar Two rules and the domestic top-up tax implementation. (Revenue) For SIs, Ireland can function as an EU base that is culturally accessible and legally integrated with the world’s largest single market. The tradeoffs are obvious and increasingly central: housing constraints, cost of living, and the fragility of any model dependent on multinational concentration. But in a world where many SIs want “EU access without EU bureaucracy at maximum intensity,” Ireland stays on the winner list.
Winner #7: South Korea (a high-tech state that is explicitly building AI governance and industrial depth). South Korea is a reminder that “winners” in the AI era are not only places with great vibes for remote workers; they are places that can make advanced hardware, scale robotics, and institutionalize AI into national strategy. In April 2025, the U.S. International Trade Administration noted South Korea passed an AI Basic Act in December 2024 to establish a legal framework to advance AI competitiveness while addressing ethical standards and trust, including institutions like an AI safety institute. (Trade.gov) South Korea’s Ministry of Science and ICT has also described the law taking effect in January 2026 and positioned it as a “new chapter in the age of AI.” (msit.go.kr) On the industrial side, South Korea remains one of the critical semiconductor pillars; Invest Korea reported semiconductors as the country’s largest export item and provided 2024 export figures (useful as a direction-of-travel indicator, not a prophecy). (InvestKorea) For an SI, South Korea is a “hard mode, high upside” base: language and cultural integration are non-trivial, but the ecosystem density in hardware, manufacturing, and increasingly AI governance makes it a winner for those who want proximity to the machinery of the AI economy, not just the software layer.
Winner #8: Taiwan (an AI and chip flywheel with exceptional upside and exceptional geopolitical risk). Taiwan is a core node of the modern world; the AI era makes that even truer because advanced compute is constrained by advanced chips, and Taiwan sits near the center of that constraint. Taiwan’s government has been articulating AI development plans; the National Development Council described a “Ten AI Initiatives Promotion Plan (2025–2028)” shaping an AI strategy to upgrade individuals, industries, and the nation. (ndc.gov.tw) There are also clear signals of “sovereign AI” thinking and governance moves; Taiwan has taken explicit security stances on AI services used in government, as Reuters reported in early 2025 when Taiwan banned government departments from using a Chinese AI service on national security grounds. (Reuters) Economically, recent analysis has linked Taiwan’s growth to AI-related exports and chips, underscoring the short-term momentum. (East Asia Forum) But Taiwan’s position comes with a structural SI warning label: it is high leverage, but it is also high tail risk. If you are an SI, Taiwan is more plausibly a place you want exposure to (supply chain, investments, partnerships) than a place you want as your only physical base—unless you intentionally build a “rapid exit” plan as part of your lifestyle architecture.
Winner #9: The “Anglosphere Safe Nodes” (Canada + New Zealand, with Australia as a conditional case). For SIs who value stability, personal safety, and “boring democracy,” the Anglosphere continues to offer some of the best-quality nodes—even if they are no longer as open as they were. Canada is explicitly adjusting immigration policy to relieve pressure on housing and services; the Government of Canada’s supplementary information for the 2026–2028 immigration levels plan states a commitment to reduce the temporary population to less than 5% of the total population by the end of 2027. (Canada) Reuters has reported that Canada is changing its Express Entry system to prioritize categories aligned with strategic sectors, while also seeking to reduce overall permanent resident inflows under Prime Minister Mark Carney’s approach. (Reuters) Canada remains a winner because it is likely to remain institutionally stable, technologically capable, and resource-rich, but it is becoming more selective, and internal political tensions around growth and immigration are visible (including provincial pushback reported by Reuters). (Reuters) New Zealand, meanwhile, has been unusually explicit in remote-work friendliness: its immigration authority states that visitor visas applied for on or after 27 January 2025 allow remote work for overseas employers or clients, with no limit on the amount of remote work you can do. (Immigration New Zealand) That is an SI-friendly signal because it reduces legal ambiguity around the “remote worker as tourist” pattern. Australia is more conditional: the Department of Home Affairs states that visitor visas do not allow work in Australia, which creates ambiguity for SIs who want to operate cleanly. (Immigration and citizenship Website) Still, Australia remains a high-quality base for those who can secure the right visa pathway and want an English-speaking, safe environment, and its skilled migration system remains active (as seen in official invitation-round reporting). (Immigration and citizenship Website) Taken together, these countries are winners because they remain desirable to the global high-agency class; the risk is that they respond to domestic strain by narrowing access, raising costs, and politicizing the SI’s relationship with the state.
Winner #10: Estonia and the “Digital State” micro-advantage (plus a small set of similarly run small countries). Estonia deserves a slot because it represents an SI-relevant innovation: the separation of “where your company lives” from “where your body lives.” Estonia’s government-backed e‑Residency program offers a digital identity that enables remote entrepreneurs to start and manage an EU company online. (e-Residency) This is not the same as tax residency or physical residency, and it does not magically eliminate bureaucracy, but it is a meaningful building block for SIs who want to modularize their life: your legal entity can be anchored in a digital-first jurisdiction even if you live elsewhere. Estonia’s e‑Residency program continues to evolve (including 2026 programming aimed at connecting founders to the local startup ecosystem). (e-Residency) The tradeoff is obvious: geography still matters in geopolitics, and Estonia sits near hard edges. So Estonia is a winner in “digital leverage,” even if it is not a universal winner in “physical safety margin.” The pattern matters beyond Estonia: small, well-run countries that are digital-first can punch above their weight by becoming nodes in the SI’s distributed life architecture.
THE TOP 10 LOSERS (COUNTRIES AND REGIONS) IN THE NEXT 5–10 YEARS
Loser #1: The Sahel and adjacent fragile belts (where climate stress meets low state capacity). If you want to predict where the next decade’s instability concentrates, you look for regions with compounding stressors: high fertility, limited fiscal capacity, weak institutions, conflict spillovers, and accelerating climate volatility. The Sahel belt—broadly Mali/Burkina Faso/Niger/Chad and surrounding zones—fits that profile. These places can still produce extraordinary people, but the state systems are structurally strained. In an AI-automated world, being “cheap labor” is less of an advantage, and regions that cannot attract capital and build high-productivity systems will see a widening gap. The result tends to be migration pressure, insurgency opportunities, and brittle politics. Even if AI lowers the cost of some services (education, translation, diagnostics), it does not magically create security, legitimacy, or infrastructure, and those are the hard constraints.
Loser #2: The Horn of Africa and conflict-linked displacement zones (Sudan as the warning flare). Global displacement is already historically high, and some of the world’s largest displacement crises are centered in African conflict zones. UNHCR reporting and global coverage have highlighted Sudan’s displacement crisis and the broader strain on humanitarian systems. (AP News) Reuters has also covered forecasts of additional displacement driven by conflicts including Sudan and Myanmar. (Reuters) The SI-relevant point is not “avoid Africa” as a continent—that’s lazy and wrong. The point is that regions where conflict is active and governance is brittle will be the least able to benefit from AI and the most likely to experience compounding instability as global attention and aid budgets fragment. Those conditions produce refugee flows and regional spillovers that stress bordering countries, including some that might otherwise be winners.
Loser #3: Afghanistan and parts of the Pakistan frontier system (high youth pressure, security volatility, limited capital absorption). This is a classic “demography meets governance” problem in a new era. Where institutions are weak and capital formation is constrained, AI doesn’t arrive as prosperity; it arrives as another layer of asymmetry. It improves the capabilities of whoever already has leverage (militias, surveillance states, black markets) faster than it improves the life of the median person. These regions also sit near major migration routes, and the SI should treat them as sources of geopolitical spillover rather than places to anchor a lifestyle base, unless you are working there with a mission and a very deliberate security posture.
Loser #4: The Levant conflict arc and fragile MENA states (where legitimacy crises and youth unemployment collide). The Middle East contains both the GCC “winner” hubs and a set of fragile states that are likely to remain under intense strain. The difference is capital + governance + stability. Where those are missing, AI and automation can remove the last remaining “labor advantage” without providing a replacement social contract. This is where extremism recruitment risk rises, where currency crises become chronic, and where migration pressure increases. Even if parts of this region stabilize, the next 5–10 years will likely remain volatile enough that SIs should treat most of it as a high-risk zone for permanent living—while recognizing that the winners in the region (UAE, parts of Saudi, Qatar) will try to firewall themselves with security and infrastructure.
Loser #5: Haiti and “fragility traps” in the Caribbean and nearby. When state capacity collapses, AI does not rebuild the state. It can help a community coordinate, and it can help diaspora networks support families, but it can’t substitute for a functional monopoly on force and basic governance. Haiti is emblematic of the fragility trap: once violence and institutional collapse cross a threshold, everything becomes expensive—energy, logistics, safety, education—and the society becomes stuck. In an AI era, that trap can deepen because the global economy becomes less dependent on low-skill labor migration as a pressure-release valve. The SI takeaway is brutal but useful: there is a category of places where you do not “bet on a turnaround” with your personal base unless you have extraordinary local knowledge and a security apparatus.
Loser #6: Venezuela and spillover zones in the northern Andes (persistent institutional and economic fracture). Some countries can muddle through crisis. Others fall into long-duration institutional fracture, where emigration becomes the primary economic strategy and the internal economy becomes a patchwork. Venezuela remains a cautionary example of what happens when institutions fail and the economy becomes structurally unstable for years. In the AI era, the risk is not merely that these places stay poor; it’s that they become sources of persistent migration and political spillover for neighboring countries, shaping security and politics across a wider region.
Loser #7: Densely populated, climate-exposed manufacturing basins that don’t upgrade fast enough (Bangladesh as the archetype). One of the darkest possibilities of the next decade is not that AI eliminates jobs everywhere equally; it’s that it disproportionately hits the countries whose export model is “labor arbitrage in global supply chains,” especially in textiles, basic manufacturing, and back-office services—while also being among the most climate-exposed and densely populated places on earth. In such environments, even modest climate shocks can become huge humanitarian problems, and automation can remove the ladder people were using to climb. The SI interpretation is that certain regions will face a double bind: their “workforce advantage” erodes at the same time their adaptation costs rise. That is a structural recipe for instability and outward migration pressure.
Loser #8: Russia and sanctioned / isolated systems (where tech progress becomes inward and brittle). Isolation is not the same as stagnation; a sanctioned state can still build powerful tech. But isolation tends to produce brittle systems: limited access to global capital, shrinking talent pools (brain drain), and increasing dependence on a narrower set of partners. In the AI era, where progress is accelerated by global research exchange, compute supply chains, and cross-border capital, isolation can become an enduring handicap. For the SI, the simplest point is practical: rule-of-law predictability and global financial connectivity matter more than ever when your life is run through digital rails. Isolation breaks rails.
Loser #9: The United Kingdom as a “mobile talent magnet” (not as a civilization—specifically as an SI-friendly fiscal regime). The UK will remain a major economy with world-class institutions in many areas, but it is plausibly a relative loser in the competition for globally mobile high-net-worth and high-agency individuals, largely because it has been tightening the fiscal regime that made it uniquely attractive to some global talent. The UK government’s technical note states that from 6 April 2025, the remittance basis for UK resident non-domiciled individuals will be abolished and replaced with a new system. (GOV.UK) Major law and accounting firms have summarized the abolition of the non-dom regime and its effective date, reflecting a meaningful regime change in the UK’s “talent magnet” policy. (Norton Rose Fulbright) This doesn’t mean the UK collapses; it means it becomes less uniquely competitive for the SI class relative to places like Switzerland, the UAE, and other hubs that actively court mobile capital and talent. In an AI era where the SI can route work globally and does not need to physically live in London to access global markets, this matters.
Loser #10: The “frontline border states” of rich regions (Southern Europe and other migration-pressure frontiers). Some places are not losers because they’re poorly run; they’re losers because geography forces them to absorb stress. Border states on major migration routes can become politically destabilized even when their institutions are decent, because the pressure is asymmetric: they face humanitarian inflows, security concerns, and domestic backlash simultaneously. With global displacement already at record levels by end of 2024, (UNHCR) and with forecasts of continued displacement, (Reuters) the next decade is likely to keep pressure on the Mediterranean frontier and other border zones. The SI implication is counterintuitive: a country can be “nice to live in” and still be a strategic loser if it becomes a pressure sink for external crises. This doesn’t mean “avoid all of Southern Europe”; it means you price in that politics and public services may become more volatile under sustained external pressure, especially if economic growth remains slow.
HOW AN SI SHOULD USE THIS MAP (WITHOUT TURNING INTO A DOOMER)
The first rule is that “winner” and “loser” are not identities; they are trajectories under stress. Countries can change course. Regions can surprise you. But the structural forces here—AI scale, automation, energy constraint, migration pressure, and institutional capacity—are not vibes. They are real constraints that will dominate the next decade’s lived experience.
The second rule is that you do not pick one place and bet your entire life on it. The SI advantage is optionality, and optionality is a system design problem. You want a multi-node life: a primary base where you can live well and operate cleanly; a secondary hub optimized for business, finance, or travel connectivity; and a contingency “exit node” that is boring, stable, and low-drama. In other words: you design your life like resilient infrastructure, because in the AI era the volatility is not only markets; it’s politics, narratives, and sudden regulatory shifts.
The third rule is that you treat compliance as part of your leverage, not an enemy of it. The temptation for high-agency people is to see taxes and regulations as obstacles to hack. That’s how you end up fragile. The real SI move is to build a clean system: transparent entity structures, audited books, legitimate residency status, and a personal security posture that assumes institutions will demand proof and traceability. The twist is that AI actually helps here: the same agentic stack that writes code and runs research can also maintain compliance, monitor policy changes, and simulate “if this rule changes, what breaks in my life?” A sovereign individual is not someone who escapes rules; it’s someone who chooses rules intentionally and builds a life that remains functional even as rules evolve.
The fourth rule is that “AI-friendly” is not always “SI-friendly.” Some countries will be extremely aggressive in deploying AI for surveillance, control, and social management. They may also be extremely safe, clean, and efficient. The SI decision is not purely economic; it’s also about what kind of relationship you want with the state. This is why the winners list is diverse: Singapore is a high-governance model, the UAE is a high-competition model, Switzerland is a high-stability model, and the U.S. is a high-variance, high-ceiling model. Different SIs will prefer different mixes.
The fifth rule is that refugee pressure and inequality are not “other people’s problems.” They are variables that will shape your daily safety, cost of living, and political environment. Even if you live inside a “winner,” a winner can destabilize if it mishandles inequality or migration. This is where the SI has to think like a realist: you want to live in places that can absorb stress without turning into permanent internal conflict.
APPENDED COMMENTARY (PHILOSOPHICAL, BUT STILL OPERATIONAL)
If you zoom out far enough, the “winner” list is not really about AI. It’s about legitimacy. AI and robotics are accelerants: they accelerate productivity, but they also accelerate inequality between those who can deploy them and those who can’t. They accelerate state capability, but they also accelerate state coercion. They accelerate migration pressure, but they also accelerate border enforcement. The next decade’s winners will be the places that can keep legitimacy intact while deploying accelerants—places that can say to their citizens, “Yes, the world is changing fast, but the system still works for you.”
One scenario worth taking seriously is the “City-State Renaissance.” Not literal city-states only, but the city-state logic: small-to-medium jurisdictions that compete fiercely for talent and capital, optimize governance, and scale compute and infrastructure faster than big countries can coordinate. Singapore and the UAE already behave this way in different styles, and the next decade might see more of it. The SI is naturally aligned with this world because SIs are modular: they can plug into a jurisdiction that matches their goals. The risk is that this produces a global archipelago of high-functioning nodes surrounded by a widening ocean of instability, which then threatens the nodes through migration, security, and moral crisis.
Another scenario is the “Fortress Regions” world: large blocs (the U.S., the EU, parts of East Asia, parts of the Gulf) harden into semi-closed systems. They invest heavily in compute and industrial capacity, regulate AI and agents, and become more selective about who gets to join. Canada’s explicit effort to reduce its temporary population share (Canada) is a mild signal in this direction: even friendly democracies may tighten not out of hostility, but out of capacity limits. In this world, being an SI means you treat residency and citizenship as long-term assets, not afterthoughts.
A third scenario is the “Automation Shock + Political Revolt” loop in the losing regions: automation reduces labor demand globally, while climate and conflict increase displacement, and then politics in winning regions becomes more volatile as citizens perceive a zero-sum competition for housing, services, and identity. You can already see how displacement at scale strains humanitarian capacity and public patience. (UNHCR) The SI question is not “will this happen?” but “how do I design a life that doesn’t get wrecked if it does?” That pushes you toward redundancy: more than one base, more than one legal status option, more than one income rail, and a personal security posture that assumes social narratives can flip quickly.
And then there is the uncomfortable question at the heart of the sovereign individual concept: if your advantage is that you can leave, what do you owe to the place you are leaving—or the place you are entering? If SIs cluster in winners, do they strengthen those societies or hollow them out into enclaves? If SIs refuse to become “data farms” for corporations, do they inadvertently become “tax farms” for states competing to monetize mobile capital? In a world where AI makes individual leverage enormous, the moral weight of individual choices grows too.
A few questions to hold as you read (and later update) this chapter: If agentic AI becomes normal, which governments will regulate it in a way that expands trust, and which will regulate it in a way that expands control? Singapore’s early move on agentic governance (Infocomm Media Development Authority) is one model; what other models emerge, and which ones become exportable norms? If displacement stays near record levels, what happens to liberal migration ideals when domestic capacity and social cohesion become binding constraints? (UNHCR) If compute becomes the new oil, do we see “compute alliances” that look like energy alliances—countries trading energy and land for data center buildout and model access? The Gulf’s investment posture and compute ambitions suggest that’s not science fiction. (Middle East Institute) Finally: if you truly become sovereign through AI, does sovereignty make you more empathetic—because you can see more, simulate more, understand more—or does it make you colder, because exit is always an option?
This is why this chapter has to be “living.” The lists will change. The signals will sharpen. Some winners will stumble; some losers will surprise. But the underlying game—governance + energy + optionality under AI acceleration—will only get more decisive.