The Futures We Can’t Avoid

We already know what's coming in the 2030s. It's not AI taking over the world.

Note to readers of this newsletter: This is a preliminary version of an article I will be publishing to a broader audience in September.
I welcome your comments. Are all these predetermined elements truly predetermined? Only if they are certain to come to pass in every plausible scenario. - Art

Satellite image of the flooding of the Ganges: NASA. Reposted from Andrew Curry, “Rereading Pierre Wack on Scenarios,” Dec 9, 2017.

Pierre Wack, the scenario pioneer who built Royal Dutch Shell’s celebrated foresight practice, sometimes explained his methods by talking about the Ganges river in northern India. If there are heavy monsoon rains over the Himalayan headwaters, you can tell with certainty that there will be a flood five days later at Allahabad, which is 650 miles downstream. Five days after that, he said, the floods would reach Benares.

“Now the people down here in Benares don’t know that this flood is on its way,” he said, “but I do. Because I’ve seen it! This is not fortune telling. This is not crystal-ball gazing. This is merely describing future implications of something that has already happened.”

Wack and his Group Planning colleagues referred to these certainties as predetermined elements: the parts of the future whose inertia exceeds any plausible intervention. Another scenario pioneer, Peter Schwartz, refers to them as inevitable surprises, because although they are certain to affect us, most of us are not emotionally prepared for them.

Most of us, peering ahead, fix on anxieties and uncertainties that may or may not happen: elections, technologies, and potential crises. We imagine what might happen, and get into the habit of thinking that our fate depends on this contingency. For instance, we pin our hopes on a particular candidate getting into office.

An alternative is to look at the predetermined elements in our world as the playing field. When we recognize the true certainties, we can leap ahead to framing our choices and modulating our expectations. For example: We know it will take a long time to mitigate the effects of the climate crisis, so we invest accordingly in renewable energy. We also know our efforts to manage artificial intelligence need to happen practically overnight, so we work to rapidly build the necessary skills.

There are two kinds of predetermined elements. The first-order trends are basic and happening now. They follow directly from events that already took place — children already born, tons of carbon already in the air, debts already incurred. The second-order ones arise from the combination of first-order forces. Their effects are less predictable, but we can’t avoid the pressures they will place on us.

Taken together, they tell us the world of the 2030s will be markedly different from today and from most predictions being made today. For system leaders, a good list of predetermined elements gives you a start on developing scenarios that help you move to a creative orientation: creating the future you want.

I do a lot of work with scenarios, particularly at New York University’s Interactive Telecommunications Program, where I teach a graduate-level course on the future of media and technology. Here is my list of predetermined elements facing us today.

First-Order Predetermined Elements

The climate crisis gets worse. Enough carbon has already accumulated that the global average temperature will cross 1.5°C above pre-industrial levels by the early 2030s. The intensity of extreme weather, and the frequency of the most damaging events, will increase accordingly. We will see more ancillary effects. For example, the cost of insuring property from weather disasters – storms, fires, and floods – will continue to rise.

Water shortages increase. Enough groundwater is being permanently depleted — fossil aquifers like the Ogallala, the North China Plain, and northwest India do not recharge on human timescales — and enough reservoirs and snowpacks are being drawn down or shrinking to make this a driving force. Chronic water shortages will affect governing regimes throughout today’s arid and semi-arid regions, including the Middle East, North Africa, South Asia, and the American Southwest.

The planet’s biological impoverishment continues. Humanity will continue to encroach on the carrying capacity of the planet. Species extinction will continue to run orders of magnitude above the natural background rate — though most people, living in cities, will register it as a fact reported rather than a loss they feel directly. There are many counter-measures emerging today, and there is a plausible future of long-term ecological recovery (though it is not predetermined). But by the 2030s, if we are not personally affected by food shortages or loss of habitat, it will be because we aren’t paying attention.

The human population balance tips. Enough of the world will have aged, enough birth rates will have fallen below replacement, and enough young people will have been born in sub-Saharan Africa that the species will be simultaneously old and young as never before. Roughly two-thirds of humanity now lives in countries below the replacement rate — China near 1.0 births per woman, South Korea around 0.75. It took less than 100 years, from 1927 to 2022, to expand from 2 billion to 8 billion humans. Now that growth is leveling off. By the mid-2030s, almost all the net growth in the young-adult workforce will be African.

There are several factors involved, all involving deep and seemingly permanent shifts in human relationships, and all permanent; despite the “culture warriors” in many countries demanding that women bear more children, there is no going back. The sociologist Alice Evans traces falling fertility to a centuries-long dissolution of patriarchal kinship: the replacement of arranged marriage by smaller, more autonomous households, and the erosion of the extended family as an economic unit. In culture after culture, women who gain autonomy defer giving birth until they are older, which means fewer children. Even if the childbearing dynamic were to change, which would contradict many other trends, it would take decades to see the aging effect reverse.

The energy system flips. In general, adoption of new technology is determined not by innovation itself but by factors related to demand: the felt need for what the technology does, the relative accessibility of the technology, the cost curve (how rapidly the price drops as use expands), and the capital already committed (including human capital). Today, those factors all favor renewable energy. Solar and wind, innovative batteries, and other manifestations of non-fossil-fuel energy are moving forward rapidly and broadly enough to be poised to become the largest single source of fuel for the global electric power system, overtaking coal and natural gas.

That doesn’t mean that fossil fuels will go away, that any one particular form of energy will corner to market, or that the cost of electricity will lessen. Demand has consistently risen to absorb new supply, and that seems predetermined to continue for some time. In this case, demand includes the computing infrastructure needed for AI servers, but also the technologies needed to manage water shortages, rising temperatures, and rising living standards around the world. Even more energy-efficient AI won’t stop demand from growing; the more energy produced, the more uses people will find for it.

Digital technologies keep maturing. This sea change in human activity began with the semiconductor 75 years ago, and its basic dynamic—continued broadening of computer use with decreased costs at a compounding rate—is predetermined to continue for at least another 10, and probably longer. Because so much technological upheaval has happened so fast, we tend to underestimate the changes still to come. We think of today’s Big Tech companies as destined to be dominant forever, but no individual platform, no matter how well-entrenched, is certain to hold its position. The explosion of large language models in 2023 is just one in a string of similar innovations, and we don’t know which will be truly transformative, in negative or positive ways: Robotics? Prosthetics? Digital fabrication? Individually designed apps? Surveillance? We only know that the pace of change and the sophistication of the tools will increase.

Medicine gets personal. A cost curve steeper than Moore’s Law is at play: sequencing a human genome cost roughly $100m in 2001 and a few hundred dollars today. The first CRISPR therapy won approval in the United States and Britain in late 2023; the mRNA platform, proven at scale during the pandemic, is being turned toward cancer. As in many technologies, the basic research takes time to translate into practical applications. That tipping point is here. Enough personalized medicine capability exists that treatment tailored to an individual’s genetic profile will be increasingly standard in oncology and rare disease by 2031 — still costly, still unevenly distributed, but no longer exotic. This is happening just in time to affect the later lives of tomorrow’s elders; perhaps not soon enough for the people in their 70s and 80s today, but well in time to transform life for many people currently in their 40s.

The fiscal vise. Enough sovereign debt has been issued, and enough of the cost of aging populations has landed on public budgets, that governments across the world will continue to face structural pressure on their finances. The debt itself is predetermined: the bonds are sold, the deficits are structural, the pension and health obligations are fixed by populations already born. What is not predetermined is whether this becomes a crisis or merely a condition, and that turns on interest rates, the general availability of investment capital, and the willingness of creditors to accept loss. Demographic logic, which the historian David Hackett Fischer traced across eight centuries, suggests that as populations shrink they need fewer houses, schools, services, and goods — so interest rates tend to fall. It’s not quite clear yet how those dynamics will play out this time.

Human capability rises overall. This trend is barely recognized, but the global generation reaching adulthood by 2031 will be the most schooled in history. Test scores may have declined in recent years in the US and other Western nations, and there may be legitimate concerns about the quality of school-based education and losing ground to artificial intelligence, but in general, the overall caliber of human skills is rising.

Second-Order: The Predetermined Combinations

The first-order elements are forces in isolation. The second-order elements are what happens when they collide. In making this list, I’ve applied a strict test. The link to two or more first-order predetermined elements must be reliable enough that the result is a pressure we can count on — even when the way people answer it remains open.

Most of these second-order surprises have, at their core, a sea change in the systems and institutions around us. They were developed for a world of young people, robust natural habitats, and prevalent internal combustion engines. It was assumed they would never have to change. Now it’s up to us to change them—and probably not through revolution. Revolution appears to be a young person’s game and there probably aren’t enough young people. Instead, we may have to change our institutions and related systems, dramatically and thoroughly, through non-violent means.

Effects of an aging world

Digital technology advances in fields where workers are scarce, not where they are cheap. One prevalent anxiety — machines coming for jobs — assumes a surplus of employees. The predetermined reality across the aging North and East is the opposite: a shortage. There will be an increase in need for help, probably in ways we don’t yet see, that combines human and digital prowess. Acemoglu and Restrepo have measured the demand: the faster a country ages, the more robots it installs. Germany, Japan, Korea, and now China automate out of necessity, not avarice. Uncertain: whether automation lifts productivity or merely fills the gap — and how the politics land where workers are still plentiful.

The cheap-labor ladder is being pulled up behind Asia. Every modern economy climbed the same ladder: cheap young workers populated the factories, and the factories built the skills and the capital. That ladder needs a reservoir of cheap young labor, and the reservoirs are draining as Asia ages and its wages rise. The last great reservoir is Africa. Uncertain: how Africa navigates this momentous transition, and whether the ladder of industrialization remains the same going forward.

Weight shifts to the old — in the ballot box and on the balance sheet. The voters of the 2030s are already alive, and they skew older every year. By plain median-voter logic, the political center of gravity tilts toward those with the largest stake in present entitlements and asset values. The same demographics concentrate accumulated wealth in older hands while the young rent and service debt. Uncertain: whether societies redistribute wealth and power across generations, let the current imbalance harden, or find a way to expand opportunity for billions of younger people while reducing the damage of today’s ecological footprint.

The dependency load rises, and “who tends the old?” becomes a central question. The ratio of people past working age to people of working age is climbing steeply, and it is locked in place: the future old are already old, and the missing workers were never born. That alone guarantees mounting fiscal and care pressure, whatever the policy. Care work, moreover, resists the productivity gains that lifted manufacturing — it is mostly hands and presence. Its human face is the fastest-growing type of household in the aging world: an old person living alone, often in a city, with no extended family to lean on. Uncertain: who supplies the need for personal care — migrants, machines, family, or no one — and whether medicine compresses the costly years or extends them.

Fewer children, and more poured into each. Gary Becker named the mechanism half a century ago: as families shrink, investment per child climbs — the quantity–quality tradeoff. China’s one-child generation is the vivid case, but the pattern is general. A world of scarcer children concentrates money, attention, and anxiety on each one. Early indicators of change: the number of young adults without cousins, and the use of social media and AI as teaching tools for human interaction. Uncertain: whether that yields a more capable generation or a more fragile, more supervised one.

Pressure to forgive and restructure debt. Enough sovereign and household debt has been issued, against populations aging enough, that the collateral cannot keep growing against a slowing economy. Fischer’s demographic logic points out that a shrinking, aging population eventually demands less capital, and the growth that once outran debt no longer does. Meanwhile, a huge amount of capital will be needed to restore Mother Earth. The pressure to mark down the debt — to forgive, inflate, or restructure — will build predictably. Uncertain: whether that pressure resolves through forgiveness, inflation, default, or grinding austerity.

Geographic changes

The largest cities humanity has ever built will rise in its hottest, poorest places. The defining human movement of the era is not the flow across the rich world’s borders. Most movement is short-range and internal: countryside to city, South to South, as a mechanizing agriculture sheds hands and a shrinking family can no longer work the land. The cohorts are born. The trajectories of Lagos, Dhaka, Kinshasa and their peers toward becoming among the largest settlements ever assembled are about as predetermined as demography gets — and they sit on some of the most heat- and flood-exposed ground on the planet. Uncertain: whether Southern megacities become engines of capability, driven by their own innovative populations, or concentrations of misery dependent on outside support.

Water becomes a problem of energy and capital — and divides the world accordingly. The aquifers are draining, and the cheapest electricity in history is arriving where the sun is strong — both locked. Together they convert water scarcity from an absolute physical limit into a question of whether you can afford the energy and the plant to manufacture water: to desalinate it, recycle it, move it. Coastal, sunny, capital-rich places increasingly can. Inland and poor places cannot. Uncertain: which regions develop their way out—and then claim leading roles in doing the same for others.

The information environment

No shared source of truth — and a rising premium on the verifiably human. When generating convincing content costs nothing, we lose what’s left of our shared and trusted picture of reality. The predetermined flip side: as the fake becomes free and infinite, the un-fakeable becomes scarce and dear. Live presence, embodied skill, provenance, a local reputation that took years to earn — these re-price upward precisely because they cannot be synthesized. Their validity is particularly important for AI systems which are seeking the credibility that comes from citing credentialed sites like Wikipedia. Uncertain: whether institutions of trust are established, or the erosion wins.

The visibility of elite impunity. Universal information colliding with extreme inequality does not in itself make the powerful more accountable. But it makes their conduct more visible. The activities of the “Epstein class” are now seen, in detail, by everyone. Other leaks and releases of information will occur. Uncertain: whether visibility converts into consequence, cynicism, or unrest, and which crimes society decides are forgivable.

AI in the environment. Driven not just by digital innovation, but also by biotech and the need for urban infrastructure, AI will reshape every city environment. Surveillance systems, facial and moement recognition, and light-and-sound engineering will bring responsive technologies outside the virtual space. It will be possible to create streets that resemble Who Framed Roger Rabbit, in which we continually encounter human-made animated systems that know us by name. Uncertain: Who will design these systems, how they will be used, and how people will respond.

The pressure every institution feels

The installed base of human error. The built world encodes the assumptions of the year it was built, and first-order change is now outrunning those assumptions faster than we replace concrete and code. Dubai built almost no stormwater drainage, on the sound assumption that it rarely rains; a warming atmosphere then delivered the rain the city was designed to exclude, most catastrophically in April 2024. Every such embedded assumption is a latent error waiting for the conditions that expose it. Uncertain: which assumptions fail, and how visibly.

The pressure to become a learning organization — even on the corrupt and the authoritarian. The most capable generation in history is reaching adulthood, while the most knowledgeable expertise on human management is available. We are all more connected, better informed, and better able to compare our lives with others (including others of the past) than ever before. The demand for adaptive, learning institutions is therefore predetermined — even where corruption is rampant and the will to reform is dormant. Uncertain: Whether the outmoded businesses and governments of our time will endure, and what would replace them if they change.

A shared incentive against catastrophe. Humanity now occupies a single niche — the world as a whole — and shares, for the first time, a common exposure. As Peter Senge puts it, “All humanity occupies a single niche: the world as a whole. We have no history of doing this successfully.” The interdependence is first-order and already here; what it produces is a common incentive not to push any catastrophe — nuclear, climatic, biological — past the point of no return. Uncertain: Whether we pay attention.

Cheap, deniable force within reach of small actors. The drone and autonomous-weapon cost curve, and its proliferation, are both locked in place. Together they put precise, deniable force in hands that never had it. The same may be true of AI-driven cyberattacks and bioattacks. Uncertain: who wields these forces, to what end, and with what technological, legal, or military restrictions.

The human cost of war rises, and people become targets rather than soldiers. Families are small and each child precious — the quantity–quality world — so low-fertility, aging societies grow acutely averse to military casualties: every soldier lost is an only child, a family line ended, and the tolerance for mass casualties that larger generations could absorb is gone. At the same time, the drone and autonomous-weapon cost curve lets a state or a faction project lethal force without putting its own young people in harm’s way. This changes the nature of war. Edward Luttwak called the casualty-averse version “post-heroic war”; the cheap drone supplies its weapon. Uncertain: whether this deters war (too few children to risk) or unleashes it (cheap, deniable force lowers the threshold), and whether any norm constrains the targeting of individuals.

The limits to certainty

The persistence of the ordinary. Against all the above sits the most underrated predetermined element of all: most people, most days, will live recognizable lives. The school, the clinic, the shop, and the family table endure because institutions change far more slowly than the forces acting on them. This is not complacency; it is the buffer that keeps the surprises survivable — and the reason that system leadership is generally local.

It’s as if we’re all driving down a treacherous highway. We notice the accidents and cars being towed off the shoulder, and the road rage as cars cut each other off. We don’t pay attention to all the drivers who stay in lane, leaving enough space between themselves and the car in front of them. Many of those drivers have experienced past accidents; they don’t want any more. If there were more of them, the road wouldn’t be nearly so scary. Uncertain: the prevailing attitudes and what it takes to bring people to a more system-oriented perspective.

Approaches for System Leaders

The discipline of scenario thinking is a discipline of attention. It tells us where to pay attention. Which predetermined elements affect us most? Which opportunities should we focus on? And which changes do we care about most urgently?

It is humbling and steadying at once: humbling because so much of the future is already decided, steadying because so much of it depends on what we do together.

The predetermined elements provide a working map. The first-order forces — the aging, the warming, the sun, the grid, the genome, the debt — are the ground on which the next decade must be built. The second-order combinations — cities, pressures, robots, possible relief — are where our work takes place.

NOTES

Predetermined elements; the Ganges metaphor: Pierre Wack, “Scenarios: Uncharted Waters Ahead,” Harvard Business Review, September–October 1985, and “Scenarios: Shooting the Rapids,” November–December 1985. The Ganges anecdote and Wack’s quotation are drawn from a filmed interview, recounted in Art Kleiner, The Age of Heretics: A History of the Radical Thinkers Who Reinvented Corporate Management, 2nd ed., Jossey-Bass, 2008, page 142.

Inevitable surprises: Peter Schwartz, Inevitable Surprises: Thinking Ahead in a Time of Turbulence, Gotham Books, 2003.

The climate crisis; 1.5°C crossed by the early 2030s: 2024 was the first calendar year more than 1.5°C above pre-industrial levels (WMO synthesis: 1.55°C); on the current emissions trajectory the long-term average is projected to pass 1.5°C in the late 2020s or early 2030s. Carbon Brief, “State of the Climate: 2024 sets a new record as the first year above 1.5C,” 10 January 2025, https://www.carbonbrief.org/state-of-the-climate-2024-sets-a-new-record-as-the-first-year-above-1-5c/. See also IPCC, Sixth Assessment Report (AR6), 2021–2023.

The cost of insuring property; property built in harm’s way: Global insured losses from natural catastrophes have grown roughly 5–7% a year in real terms over three decades, driven chiefly by “more valuable property being built in harm’s way.” Swiss Re Institute, sigma 1/2025, “Natural catastrophes: insured losses on trend to USD 145 billion in 2025,” https://www.swissre.com/institute/research/sigma-research/sigma-2025-01-natural-catastrophes-trend.html.

Groundwater permanently depleted; fossil aquifers: Groundwater is declining in 71% of the world’s aquifers, with declines accelerating in arid and semi-arid farmland. Scott Jasechko et al., “Rapid groundwater decline and some cases of recovery in aquifers globally,” Nature 625 (2024): 715–721, https://doi.org/10.1038/s41586-023-06879-8. (The Ogallala is the clearest case of a fossil aquifer that does not recharge on human timescales; the North China Plain has since seen a recent engineered, diversion-driven partial recovery.)

Encroaching on the carrying capacity of the planet: The Stockholm Resilience Centre frames this as “planetary boundaries” rather than “carrying capacity.” As of the 2025 Planetary Health Check, seven of nine planetary boundaries are transgressed, with ocean acidification newly added in 2025. “Seven of nine planetary boundaries now breached,” Stockholm University, 25 September 2025, https://www.su.se/english/news/articles/2025-09-25-seven-of-nine-planetary-boundaries-now-breached; framework: Katherine Richardson et al., Science Advances (2023), and Johan Rockström et al., Nature (2009), https://www.stockholmresilience.org/research/planetary-boundaries.html.

Species extinction above the background rate: The current global rate of species extinction is “tens to hundreds of times higher than the average over the last 10 million years,” with around one million species threatened. IPBES, Global Assessment Report on Biodiversity and Ecosystem Services, 2019, https://www.ipbes.net/news/Media-Release-Global-Assessment.

Two-thirds of humanity below replacement: Population Reference Bureau, “Beyond the Headlines: What’s Really Happening With Global Fertility?” October 2025, https://www.prb.org/resource/beyond-the-headlines-whats-really-happening-with-global-fertility/; United Nations, World Fertility 2024, https://www.un.org/development/desa/pd.

China near 1.0; South Korea around 0.75: United Nations, World Population Prospects 2024 (2025 estimates): China at about 1.0 and South Korea at roughly 0.72–0.75 births per woman; China, the Republic of Korea, Singapore, and Ukraine are below 1.0. United Nations, World Fertility 2024.

2 billion in 1927 to 8 billion in 2022; growth leveling off: World population reached 8.2 billion in 2024 and is projected to peak near 10.3 billion in the mid-2080s before declining. United Nations, World Population Prospects 2024, https://population.un.org/wpp/.

Net growth in the young-adult workforce will be African: United Nations, World Population Prospects 2024; sub-Saharan Africa accounts for nearly all projected net growth in the world’s working-age population through mid-century.

Falling fertility and the dissolution of patriarchal kinship: Alice Evans, The Great Gender Divergence (Substack), https://www.ggd.world/, and her forthcoming book of the same title (Princeton University Press).

Renewables become the largest single source of electricity: Renewables are set to overtake coal as the largest source of global electricity generation at the end of 2025, their share rising from about 32% in 2024 toward 43% by 2030. IEA, Renewables 2025, https://www.iea.org/reports/renewables-2025/renewable-electricity; IEA, Electricity 2026, https://www.iea.org/reports/electricity-2026.

Cost curves and “the cheapest electricity in history”: The IEA described the best solar projects as offering “the cheapest electricity in history.” IEA, World Energy Outlook 2020, https://www.iea.org/reports/world-energy-outlook-2020.

A cost curve steeper than Moore’s Law; genome sequencing: Sequencing a human genome cost about $95 million in 2001 and a few hundred dollars today, a decline steeper than Moore’s Law. National Human Genome Research Institute, “DNA Sequencing Costs: Data,” https://www.genome.gov/about-genomics/fact-sheets/DNA-Sequencing-Costs-Data.

The first CRISPR therapy approved in the US and Britain, late 2023: Casgevy (exagamglogene autotemcel) was approved by the UK MHRA on 16 November 2023 and by the US FDA on 8 December 2023. “First Regulatory Approvals for CRISPR-Cas9 Therapeutic Gene Editing for Sickle Cell Disease and Transfusion-Dependent β-Thalassemia,” editorial, 2024, https://pubmed.ncbi.nlm.nih.gov/38425279/.

The mRNA platform turned toward cancer: Jeffrey S. Weber et al., “Individualised neoantigen therapy mRNA-4157 (V940) plus pembrolizumab versus pembrolizumab monotherapy in resected melanoma (KEYNOTE-942): a randomised, phase 2b study,” The Lancet 403 (2024): 632–644, https://doi.org/10.1016/S0140-6736(23)02268-7.

Demographic logic traced across eight centuries: David Hackett Fischer, The Great Wave: Price Revolutions and the Rhythm of History, Oxford University Press, 1996.

The most schooled generation in history (and the learning caveat): Mean years of schooling have risen steadily worldwide: Our World in Data, “Global Education,” https://ourworldindata.org/global-education. On the quality caveat the draft acknowledges, the World Bank estimates about 70% of 10-year-olds in low- and middle-income countries cannot read and understand a simple text. World Bank, The State of Global Learning Poverty: 2022 Update, https://www.worldbank.org/en/topic/education/publication/state-of-global-learning-poverty.

Automation arrives where workers are scarce, not cheap: Daron Acemoglu and Pascual Restrepo, “Demographics and Automation,” The Review of Economic Studies 89, no. 1 (2022): 1–44, https://doi.org/10.1093/restud/rdab031.

The cheap-labor ladder; deindustrialization at lower incomes: Dani Rodrik, “Premature Deindustrialization,” Journal of Economic Growth 21, no. 1 (2016): 1–33, https://doi.org/10.1007/s10887-015-9122-3.

Care work resists the productivity gains that lifted manufacturing: William J. Baumol, The Cost Disease: Why Computers Get Cheaper and Health Care Doesn’t, Yale University Press, 2012. Old-age dependency ratios from United Nations, World Population Prospects 2024.

Fewer children, more invested in each (quantity–quality): Gary S. Becker and H. Gregg Lewis, “On the Interaction between the Quantity and Quality of Children,” Journal of Political Economy 81, no. 2 (1973): S279–S288; Gary S. Becker, A Treatise on the Family, Harvard University Press, 1981.

The largest cities ever built, in the hottest, poorest places: United Nations, World Urbanization Prospects 2025, https://population.un.org/wup/; Lagos, Kinshasa, and Dhaka rank among the world’s largest projected urban agglomerations.

The installed base of human error; the Dubai floods: Dubai recorded its heaviest rainfall since records began in 1949 on 16 April 2024; its limited stormwater drainage reflects an infrastructure designed for an almost rainless climate. NASA Earth Observatory, “Deluge in the United Arab Emirates,” https://earthobservatory.nasa.gov/images/152703/deluge-in-the-united-arab-emirates; Associated Press, “UAE struggles to recover after heaviest recorded rainfall,” April 2024.

A single planetary niche (Senge): Peter M. Senge, “Introduction to the Living Fieldbook Project,” forthcoming. The Anthropocene framing draws on Colin Waters et al., “The Anthropocene Is Functionally and Stratigraphically Distinct from the Holocene,” Science 351, no. 6269 (2016), https://doi.org/10.1126/science.aad2622.

The human cost of war; post-heroic warfare: Edward N. Luttwak, “Toward Post-Heroic Warfare,” Foreign Affairs 74, no. 3 (May/June 1995): 109–122.

A few elements rest on reasoning from the first-order facts above rather than on one citable statistic, and are sourced only by the elements they combine: the shift of political and economic weight to older cohorts (standard median-voter analysis — Anthony Downs, An Economic Theory of Democracy, Harper, 1957 — plus higher turnout among older voters); the erosion of a shared source of truth and the rising premium on the verifiably human; AI moving into the physical environment; cheap, deniable force within reach of small actors; and the persistence of the ordinary.