On the Future

Martin Rees argues that the twenty-first century is uniquely consequential in Earth's 4.5-billion-year history because one species, humans, now holds the planet's future in its hands. Writing as a scientist, a citizen, and a self-described anxious member of the human race, Rees surveys the threats and opportunities facing humanity, contending that science and technology can secure a bright future but only if deployed wisely and governed effectively.

In a preface written for the paperback edition, Rees frames the COVID-19 pandemic as a demonstration of the book's central tensions. The crisis revealed both the fragility of an interconnected world and the power of well-directed science, as advances in immunology enabled rapid vaccine development. He warns against preparing only for the last crisis: Other scenarios, including massive cyberattacks, bioterrorism, cascading infrastructure failure, and accidental nuclear war, grow in likelihood year by year. He calls for new international institutions and long-term planning, urging the older generation to strive to be "good ancestors."

The book opens with a thought experiment. Hypothetical aliens observing Earth over its entire history would see agriculture, urbanization, rising atmospheric carbon dioxide, and rockets escaping the biosphere as abrupt disruptions in the planet's final sliver of geological time. Rees asks whether the next century will bring ecological stabilization and an expansion of life beyond Earth, or a final spasm followed by silence. He notes a paradox: Despite vastly expanded scientific understanding, the timescale on which we can sensibly plan has shortened, because social and technical change now outstrips the slow rhythms of biology and geology.

Chapter 1 surveys the human-induced hazards of what some geologists call the Anthropocene, the current epoch in which human activity dominates Earth's ecology. Natural threats like asteroid impacts are quantifiable and rare, but the far greater hazards are those humans create. Rees recounts how close the world came to nuclear catastrophe during the Cold War: During the 1962 Cuban missile crisis, President Kennedy estimated the odds of nuclear war at between one in three and even. Soviet submarine officer Vasili Arkhipov refused to authorize a nuclear torpedo launch, averting potential catastrophic escalation, and in 1983 Russian Air Force officer Stanislav Petrov chose on a hunch to ignore a false alarm indicating a US missile strike. Though arsenals have shrunk roughly fivefold, nine nuclear powers now exist, and Rees argues that a near-existential nuclear threat is merely in abeyance.

Rees examines the interlinked pressures of population growth, biodiversity loss, and climate change. World population has risen from roughly 3.5 billion fifty years ago to 7.6 billion and is projected to reach around nine billion by 2050. He argues that no single carrying-capacity figure is meaningful without specifying lifestyle assumptions, and that modern agriculture could plausibly feed nine billion through "sustainable intensification," an approach that boosts food output while limiting waste, water use, and ecological damage. He introduces the concept of "planetary boundaries"—the ecological thresholds limiting how heavily human activity can burden Earth's natural systems—warning that crossing such tipping points could irreversibly impoverish the biosphere. On climate, Rees explains the uncontroversial greenhouse science but notes that uncertain feedback processes make total warming projections difficult to pin down. He identifies three mitigation measures: improving energy efficiency through circular-economy principles, meaning reuse-and-repair-focused design; cutting subsidiary greenhouse gases like methane and black carbon; and expanding research into low-carbon energy. He predicts that political efforts to decarbonize will not gain sufficient traction, and that if climate sensitivity, the degree of warming that follows a given CO₂ increase, proves high, panic measures such as geoengineering—deliberate large-scale interventions in Earth's climate system, for instance by seeding the upper atmosphere with reflective aerosols—could be pursued despite enormous complications.

Chapter 2 turns to transformative technologies. In biotech, the cost of sequencing the human genome has plummeted from three billion dollars to under one thousand, and gene-editing tools like CRISPR/Cas9 allow precise modifications to DNA, raising both medical promise and ethical alarm. Unlike nuclear weapons, which require large special-purpose facilities, biotech involves small-scale, dual-use equipment accessible to many, making worldwide regulation nearly impossible. Controversial gain-of-function experiments, which deliberately enhance a pathogen's transmissibility or virulence, and the 2018 synthesis of horsepox illustrate the risks. Rees reports his own 2003 prediction of a 50 percent chance that bio error or bioterror would cause a million deaths by 2020, and notes that Steven Pinker, known for arguing that violence has declined historically, bet against him. Rees contends that Pinker's optimism underestimates the asymmetric risk of rare but extreme events.

In cybertechnology and artificial intelligence, Rees highlights DeepMind's AlphaGo program, which beat the world champion of Go in 2016 by learning from games rather than being programmed by experts. AI excels at analyzing vast data sets, but Rees raises concerns about opaque decision-making, ubiquitous data collection, and the displacement of manufacturing and routine white-collar jobs. He warns that autonomous weapons capable of identifying and killing individuals through facial recognition represent a deeply troubling development. Looking further ahead, he catalogs existential risks from cascading network failures to a powerful AI pursuing goals hostile to humanity, and introduces philosopher Derek Parfit's argument that human extinction would be incomparably worse than even a 90 percent die-off because it would foreclose the existence of billions of future people. He concludes that "the unfamiliar is not the same as the improbable."

Chapter 3 places humanity in a cosmic perspective. Modern telescopes have revealed that most stars are orbited by planets, and statistical extrapolation suggests the Milky Way harbors around a billion planets where liquid water could exist. Whether any harbor life remains unknown. On spaceflight, Rees argues that the future lies with private companies like SpaceX, the rocket venture led by Elon Musk, and Blue Origin, the spaceflight company backed by Jeff Bezos, rather than national agencies. He strongly disagrees with Musk and the late physicist Stephen Hawking about building large Martian communities, calling it "a dangerous delusion to think that space offers an escape from Earth's problems." Instead, he argues that space-faring adventurers, beyond the reach of terrestrial regulators, will spearhead the posthuman era by using genetic and cyborg technologies to adapt to hostile environments, potentially transitioning to fully inorganic intelligence. On alien intelligence, Rees contends that any extraterrestrial signal would most likely come from electronic rather than organic beings, since organic technological civilization is generically a brief interlude before machines take over.

Chapter 4 explores the limits of scientific understanding. Rees uses John Conway's Game of Life, in which simple rules iterated repeatedly produce complex patterns, to illustrate how complexity emerges from basic principles. He speculates about the multiverse, the idea that our big bang may be one among many governed by different physical laws, and asks whether some features of reality may be permanently beyond human comprehension. On science and religion, Rees states that he does not believe in God but shares a sense of wonder with many who do, describing himself as a "practising but unbelieving Christian."

Chapter 5 argues that all citizens need enough scientific literacy to assess risks and resist misleading claims. Scientists bear special obligations to foster benign applications of their work and engage the public, as exemplified by the atomic scientists who built nuclear weapons and then advocated arms control. Nations may need to cede sovereignty to international organizations to address threats no single country can manage. The book concludes that no scientific impediment prevents a sustainable world where all enjoy a better lifestyle, but the gap between technological possibility and political reality engenders deep pessimism. Rees calls for thinking globally, rationally, and long-term, "empowered by twenty-first-century technology but guided by values that science alone can't provide."