Simulated Worlds and Rotating Spacecraft

In celebration of Earth Month, MoMA’s Emilio Ambasz Institute for the Joint Study of the Built and the Natural Environment presents the inaugural entry in the Ambasz Essays, a new series of commissioned articles that explore how the built and natural environment intersect with topics like the climate crisis, resource extraction, environmental justice, and race and indigeneity.

It’s no secret that we aren’t great at predicting the future. In 1966, Time magazine declared that “remote shopping, while entirely feasible, will flop.” Four years later, the famous MIT computer scientist Marvin Minsky told Life magazine that within “three to eight years we will have a machine with the general intelligence of an average human being.” And Marty Copper, called the “father of the (handheld) cell phone,” confidently stated in 1981 that “cellular phones will absolutely not replace local wire systems.”

Renaissance Man. 2023. Dalle-2 AI-generated image

“Imagine hopping in a time machine and setting the destination for somewhere in Europe circa 1700. Could anyone at the time have prophesied the automobile, nuclear missiles, or Twitter?”

The difficulty of prediction hasn’t dissuaded some futurists from making bold claims about what things could—and should—look like thousands, millions, and even billions of years from now. These days, one of the most popular futurological visions in Silicon Valley and among tech elites—a view embraced by Elon Musk himself—is called longtermism. Though its roots go back to the 1980s, it wasn’t until the past two decades that longtermism morphed into a cohesive worldview, thanks in part to the work of Nick Bostrom, a controversial philosopher at Oxford University. Two predictions are central to this vision: first, that humanity will spread beyond Earth, leaping from one solar system to another, expanding to other galaxies, and eventually planting our flag throughout most of the accessible universe. Second, that our descendants will use material resources floating in outer space, such as exoplanets, to build enormous supercomputers powered by “Dyson swarms,” i.e., megastructures that would envelop stars (like our Sun) to harvest most of their energy. These supercomputers would then run detailed virtual-reality worlds in which large numbers of “digital people”—conscious beings like you and me—would live out their lives.

The Milky Way galaxy imaged using infrared light, 2019

In the longtermist view, the more people there are in the future, the better the future will be—a moral claim that’s contentious among philosophers—and hence these aren’t just predictions but imperatives. They are goals that humanity should set for itself; they constitute part of what it means to “fulfill our long-term potential” in the cosmos, according to Toby Ord, a leading longtermist who is also at Oxford.

Most people find this vision of the distant future rather bizarre, if not off-putting. I share that sentiment. However, given that many powerful figures—such as tech billionaires—are sympathetic with longtermism, it’s entirely possible that this is what our future will end up looking like. If so, the implications are profound. Consider first that virtual reality—not just the sort of virtual reality you might plug into through VR goggles, but a “reality” in which one is fully embedded as a digital being—would constitute the ultimate built environment. Within this environment, there might be wild animals, ecosystems, rivers, and forests, yet even these “natural” parts of the simulated world would be artificial. The distinction between natural and built environments might still make sense and be useful to people inhabiting this world, though both would arise from 1s and 0s in some underlying computer program. (For all we know, our current world is a simulation, in which case everything we call “natural” is really artificial anyway.)

In a simulated reality, there is almost no limit to the possibilities that could be explored. Consider that our world is defined by four fundamental forces, including electromagnetic and gravitational forces, and a rather short list of fundamental physical constants, like the speed of light. But there is no reason we couldn’t alter these to create wildly exotic milieus radically different from our own. Put differently, the universe itself would fall within the architect’s purview: rather than buildings designed for a particular environment, which is largely beyond our control in this world, the environment itself could be designed in tandem with whatever structures it might contain. If, for example, gravity were made weaker in a simulated world than it is on Earth, it could become possible to build skyscrapers much taller than the 2,700-foot Burj Khalifa—maybe 50 or 100 times taller. Altering other physical forces and constants could open the door to even more alien options. While buildings today are generally static over time—the Eiffel Tower, for example, retains the same shape it had when constructed back in 1889—simulated worlds could enable buildings to morph and evolve dynamically over time. An entire virtual city could cycle through different layouts, appearances, and styles, perhaps offering a unique combination for each day of the week.

The situation becomes even more mind-boggling when one realizes that inhabitants of simulated worlds might be radically different from us. This matters because the suitability of the built environment depends crucially on the beings that inhabit it. In our present world, each of us has our own unique psychological history. Our memories are entirely our own, even if formed during experiences with others, and the personality traits that define us are relatively stable across time. This needn’t be the case in a simulated world. As the sociologist James Hughes writes, digital beings would “be able to copy, share, and sell [their] memories, beliefs, skills, and experiences.” Personalities would “begin to bleed and blur,” and indeed “the most dramatic challenges to our social and philosophic world will probably come from hive minds and distributed selves.” What sort of world would an architect build for such beings? Indeed, what would it even mean to be an “architect” if one’s knowledge and skills could be copy-and-pasted from other minds floating in this digital soup of existence?

O’Neill Cylinder. 1974. Painting by Rick Guidice, NASA Ames Research Center

Long before we construct huge computer simulations on which to run high-resolution VR worlds, however, our descendants would likely establish colonies on other planets, such as Mars, and build enormous spacecraft. These, too, would present exciting new opportunities for radically reimagining the environments we inhabit. Consider the case of O’Neill cylinders, an idea first proposed in the 1970s by Gerard O’Neill, a Princeton physicist whose courses happened to be attended by a young Jeff Bezos, now a vocal advocate for O’Neill’s vision. Imagine a giant cylinder with a diameter of about five miles, or 26,400 feet, rotating roughly 28 times per hour to produce a “centrifugal” force that approximates the “pull” of gravity here on Earth. Up to 20 million people could live inside it, surrounded by mountains, forests, lakes, and rivers, and indeed O’Neill pictured residents of these cylinders enjoying sports like mountain climbing and skiing. In this case, rather than the natural environment encompassing the built environment, as is the case on Earth, the “natural” environment would be subsumed within the built one: a self-contained oasis amid the infinite vastness of space. Everything within the cylinder would be regulated by human design: the amount of sunlight that enters the cylinder each day, the composition of the atmosphere (oxygen, nitrogen, etc.), the various cycles that comprise the ebb-and-flow of ecological processes. O’Neill cylinders would constitute the ultimate ecotechnological system, with natural and artificial systems interwoven in a symbiotic relationship of mutual self-dependence: plants and animals would depend on people, and people would depend on these plants and animals.

Just as intriguing are what cities and towns in these cylinders might look like. If the cylinder’s diameter is 26,400, its radius would be 13,200 feet—nearly five times greater than the height of the Burj Khalifa. At the center of this cylinder, 13,200 feet above the ground in all directions, you’d be weightless, just as you would be if you were to tunnel to the very center of Earth (and somehow survived the extreme heat and the weight crushing you from all directions). In fact, the “pull” of “gravity” from the cylinder’s rotation would decrease linearly with altitude, which means if you were positioned halfway between the cylinder’s center and the ground beneath, you’d weigh half as much. This is wildly different than our experience of gravity on Earth: if you drive up a mountain until you’re 6,600 feet above sea level (that’s half of 13,200), you won’t cut your weight in half; if you drive up another 6,600 feet, you won’t be completely weightless. Indeed, it’s impossible to notice the reduction in gravity’s “pull” even when you’re cruising in an airplane at 42,000 feet!

Consequently, buildings within O’Neill cylinders could take on wildly different shapes than here on the home planet, third rock from the sun. The taller they are, the less each additional floor will weigh, and hence one could imagine large structures occupying the cylinder’s center, in microgravity, held in place by structures extending in various directions like the roots of a mangrove tree. Imagine going to work, hopping in the elevator, and getting out on a floor high above the cylinder's ground level—where suddenly you can easily jump three or four feet in the air. During a 30-minute afternoon break, you take the elevator further up, until you’re free-floating in microgravity and peering down at your village between two mountains from a height of 13,200 feet. Indeed, O’Neill himself imagined that residents of these cylinders might even invent entirely new types of sports—which would take advantage of the peculiar features of the spinning cylinder that our spacefaring descendants would call home—such as “three-dimensional soccer.”

“The future ain’t what it used to be,” as Gary Goshgarian, an English professor at Northeastern University, told the Boston Globe in 1973. One way to interpret this notion is that our visions of the future keep changing over time: the 1927 film Metropolis imagined a dystopian world full of robots and machines, while people in the 1950s anticipated flying cars and docile mechanical housemaids. Today, the longtermist vision has captured the attention of many powerful actors in the tech sector—though it’s also built on predictions about the technological capabilities of future humans to construct O’Neill cylinders, Dyson swarms, planet-sized supercomputers, and the like. For all we know, none of this will be possible; even if it is technically feasible, scholars like Daniel Deudney have made compelling cases that colonizing our solar system itself will trigger catastrophic wars that in all likelihood will result in total human extinction.

Nonetheless, the momentum at present points to a future among the heavens, and if this comes to pass the implications would be enormous. One thing is for sure: the world in which we live is increasingly shaped by human intervention and increasingly “artificial”; if this trend continues, our descendants may well find themselves in a Matrix of our own making.

Heinz Schulz-Neudamm. Poster for the film Metropolis. 1926

Émile P. Torres is a philosopher and historian whose work focuses on the ethical implications and potential causes of human extinction. Their forthcoming book is Human Extinction: A History of the Science and Ethics of Annihilation.