The Singularity Is Nearer: When We Merge with AI

by Ray Kurzweil

AI-driven advances in materials science will make solar photovoltaic electricity extremely cheap, while robotic resource extraction and autonomous electric vehicles will bring the costs of raw materials much lower. With cheap energy and materials, and automation increasingly replacing human labor outright, prices will fall substantially. In time, such effects will cover so much of the economy that we’ll be able to eliminate much of the scarcity that presently holds people back. As a result, in the 2030s it will be relatively inexpensive to live at a level that is considered luxurious today.

Before the release of the iPhone in 2007, there was no app economy to speak of. In 2008 there were fewer than 100,000 iOS apps available; this had rocketed up to around 4.5 million by 2017.[126] On Android, the growth was just as dramatic. In December 2009, there were around 16,000 mobile apps available in the Google Play Store.[127] As of March 2023, there were 2.6 million.[128] That is a more than 160-fold increase in thirteen years.

Although it is not without its limitations, the so-called gig economy often allows people more flexibility, autonomy, and leisure time than their previous options. Maximizing the quality of these opportunities is one strategy for how to help workers as automation trends accelerate and disrupt traditional workplaces.

Yep, Kurzweil gets it. "Just be an Uber driver, bro!"

For example, when I was growing up, there were only three television stations available: ABC, NBC, and CBS. Because everyone was watching such a limited selection of programs, the networks had to create content that would be popular with the widest possible demographics. In order to be successful, shows had to appeal to men and women, children and parents, blue collar and white collar. Ideas with a strong but narrower appeal, like absurdist comedy, paranormal drama, or science fiction, did not have an easy path to commercial viability. Many people now forget that Star Trek, the most influential science fiction series in history, was canceled after just three seasons.[133]

These capabilities will become even more integrated with our lives throughout the 2020s. Search will transform from the familiar paradigm of text strings and link pages into a seamless and intuitive question-answering capability. Real-time translation between any pair of languages will become smooth and accurate, breaking down the language barriers that divide us. Augmented reality will be projected constantly onto our retinas from our glasses and contact lenses. It will also resonate in our ears and ultimately harness our other senses as well. Most of its functions and information will not be explicitly requested, but our ever present AI assistants will anticipate our needs by watching and listening in on our activities. In the 2030s, medical nanorobots will begin to integrate these brain extensions directly into our nervous system.

Remember, though, that technological abundance doesn’t automatically benefit everyone equally at once. For example, in 2022, $1.00 could buy more than 50,000 times as much computing power as it could in 2000 (adjusted for overall inflation).[153] By contrast, according to official statistics, $1.00 in 2022 could buy only about 81 percent of the health care it could buy in 2000 (adjusted for overall inflation).[154] And although some medical treatments, like cancer immunotherapies, got qualitatively better in that time, most health-care expenditures, such as hospital stays and X-rays, remained roughly the same. Thus, people who spend a lot of their money on computers, like students and young people, got a lot of benefit from falling computing prices. Meanwhile, those who spend a big part of their income on health care, like the elderly and those with chronic illnesses, may have even been worse off overall.

People often say that it is death and the brevity of our existence that gives meaning to life. But my view, rather, is that this perspective is an attempt to rationalize the tragedy of death as a good thing. The reality is that the death of a loved one literally robs us of part of ourselves. The neocortical modules that were wired to interact with and enjoy the company of that person now generate loss, emptiness, and pain. Death takes from us all the things that in my view give life meaning—skills, experiences, memories, and relationships. It prevents us from enjoying more of the transcendent moments that define us—producing and appreciating creative works, expressing a loving sentiment, sharing humor.

Imagine trying to explain music, literature, a YouTube video, or a joke to primates that missed out on the neocortical expansion that the large foreheads of hominids made possible two million years ago. This analogy helps us to at least glimpse how, when we digitally augment our neocortex starting sometime in the 2030s, we will be able to create meaningful expressions that we cannot imagine or understand today.

People actually adapt very quickly to change, especially change for the better. In the late 1980s, when the internet was still mainly limited to universities and governments, I predicted that a vast worldwide network for communication and information sharing would eventually be available to everyone, even schoolchildren, by the late 1990s.[163] I also predicted the advent of mobile devices that people would use to harness this network by the early twenty-first century.[164] Such predictions seemed daunting and disruptive (not to mention unlikely) when I made them, but they actually came true, and those technologies were very rapidly adopted and accepted. As just one example, the entire app economy barely existed a decade and a half ago, yet it is now so deeply established that people can hardly remember when it wasn’t around.

Stanford research found that an estimated 39 percent of American heterosexual couples in 2017 had met online—many of those through mobile apps like Tinder and Hinge.[165] That means that many of the children now in elementary school exist only because of a technology that’s just a few years older than they are. When you take a step back to look at these changes in perspective, the speed with which apps have affected society is truly amazing.

Imagine the proportion now!

Imagining a world where humans are largely in competition with AI-powered machines is the wrong way of thinking about the future. To illustrate this, imagine a time traveler with a 2024 smartphone going back to 1924.[166] This person’s intelligence would seem truly superhuman to the people of Calvin Coolidge’s day. They could do advanced math effortlessly, translate any major language passably well, play chess better than any grandmaster, and command a whole Wikipedia’s worth of facts. To the people of 1924, it would seem obvious that the time traveler’s capabilities were radically enhanced by the phone. But to us in the 2020s it’s easy to lose this perspective. We don’t feel augmented. In a similar way, we will harness the advances of 2030 and 2040 to seamlessly augment our own capabilities—and this will feel even more natural as our brains interface directly with computers.

The most important reason for optimism about the coming social transitions is that growing material abundance will lower the incentives for violence. When people don’t have the necessities of life, or when crime is already high, citizens can feel that they have nothing to lose by violence. But the same technologies that will cause these social disruptions will also be making food, housing, transportation, and health care much cheaper. And crime will likely keep falling through a combination of better education, smarter policing, and a reduction in environmental toxins like lead that damage people’s brains. When people feel that they have long, safe lives ahead of them, they have stronger incentives to work out their differences through politics instead of risking everything by resorting to force.

Like most white liberals, Kurzweil lives in a self-imposed bubble. One meeting with a Muslim would set him straight.

Perhaps the most important class of problems at present is designing treatments for emerging viral threats. This challenge is like finding which key will open a given virus’s chemical lock—from a pile of keys that could fill a swimming pool. A human researcher using her own knowledge and cognitive skills might be able to identify a few dozen molecules with potential to treat the disease, but the actual number of possibly relevant molecules is generally in the trillions.[3] When these are sifted through, most will obviously be inappropriate and won’t warrant full simulation, but billions of possibilities may warrant a more robust computational examination. At the other extreme, the space of physically possible potential drug molecules has been estimated to contain some one million billion billion billion billion billion billion possibilities![4] However one frames the exact number, AI now lets scientists sort through that gigantic pile to focus on those keys most likely to fit for a given virus.

In our current paradigm, once we have a potentially feasible disease-fighting agent, we can organize a few dozen or a few hundred human subjects and then test them in clinical trials over the course of months or years at a cost of tens or hundreds of millions of dollars. Very often this first option is not an ideal treatment: it requires exploration of alternatives, which will also take a few years to test. Not much further progress can be made until those results are available. The US regulatory process involves three main phases of clinical trials, and according to a recent MIT study, only 13.8 percent of candidate drugs make it all the way through to FDA approval.[5] The ultimate result is a process that typically takes a decade to bring a new drug to market, at an average cost estimated between $1.3 billion and $2.6 billion.[6]

But by far the most important application of AI to medicine in 2020 was the key role it played in designing safe and effective COVID-19 vaccines in record time. On January 11, 2020, Chinese authorities released the virus’s genetic sequence.[11] Moderna scientists got to work with powerful machine-learning tools that analyzed what vaccine would work best against it, and just two days later they had created the sequence for its mRNA vaccine.[12] On February 7 the first clinical batch was produced. After preliminary testing, it was sent to the National Institutes of Health on February 24. And on March 16—just sixty-three days after sequence selection—the first dose went into a trial participant’s arm. Before the pandemic, vaccines typically took five to ten years to develop. Achieving this breakthrough so quickly surely saved millions of lives.

All the applications I’ve described are instances of a much more fundamental challenge in biology: predicting how proteins fold. The DNA instructions in our genome produce sequences of amino acids, which fold up into a protein whose three-dimensional features largely control how the protein actually works. Our bodies are mostly made of proteins, so understanding the relationship between their composition and function is key to developing new medicines and curing disease. Unfortunately, humans have had a fairly low accuracy rate at predicting protein folding, as the complexity involved defies any single easy-to-conceptualize rule. Thus, discoveries still depend on luck and laborious effort, and optimal solutions may remain undiscovered. This has long been one of the main obstacles to achieving new pharmaceutical breakthroughs.[14] This is where the pattern recognition capabilities of AI offer a profound advantage. In 2018 Alphabet’s DeepMind created a program called AlphaFold, which competed against the leading protein-folding predictors, including both human scientists and earlier software-driven approaches.[15] DeepMind did not use the usual method of drawing on a catalog of protein shapes to be used as models. Like AlphaGo Zero, it dispensed with established human knowledge. AlphaFold placed a prominent first out of ninety-eight competing programs, having accurately predicted twenty-five out of forty-three proteins, whereas the second-place competitor got only three out ...

AI will scale up to modeling ever larger systems in simulation—from proteins to protein complexes, organelles, cells, tissues, and whole organs. Doing so will enable us to cure diseases whose complexity puts them out of the reach of today’s medicine. For example, the past decade has seen the introduction of many promising cancer treatments, including immunotherapies like CAR-T, BiTEs, and immune checkpoint inhibitors.[20] These have saved thousands of lives, but they frequently still fail because cancers learn to resist them. Often this involves tumors altering their local environment in ways we can’t fully understand with current techniques.[21] When AI can robustly simulate the tumor and its microenvironment, though, we’ll be able to tailor therapies to overcome this resistance. Likewise, such neurodegenerative diseases as Alzheimer’s and Parkinson’s involve subtle, complex processes that cause misfolded proteins to build up in the brain and inflict harm.[22] Because it’s impossible to study these effects thoroughly in a living brain, research has been extremely slow and difficult. With AI simulations we’ll be able to understand their root causes and treat patients effectively long before they become debilitated. Those same brain-simulation tools will also let us achieve breakthroughs for mental health disorders, which are expected to affect more than half the US population at some point in their lives.[23]

It is remarkable that biology has created a creature as elaborate as a human being, one with both the intellectual dexterity and the physical coordination (e.g., opposable thumbs) to enable technology. However, we are far from optimal, especially with regard to thinking. As Hans Moravec argued back in 1988, when contemplating the implications of technological progress, no matter how much we fine-tune our DNA-based biology, our flesh-and-blood systems will be at a disadvantage relative to our purpose-engineered creations.[45] As writer Peter Weibel put it, Moravec understood that in this regard humans can only be “second-class robots.”[46] This means that even if we work at optimizing and perfecting what our biological brains are capable of, they will be billions of times slower and far less capable of what a fully engineered body will be able to achieve.

Because the wet biological environment inside our skulls is (at least at the molecular level) a very turbulent place, any single neuron may die or simply fail to fire at the correct instant. If human cognition depended heavily on the performance of any single neuron, it would be very unreliable. But when many neurons work together in parallel, the “noise” gets canceled out and we’re able to think just fine.

The interior of a computer chip is much more clean and stable than brain tissue, so all that parallelism won’t be necessary. This will allow for more efficient computation, so it’s plausible that a mind could be simulated with even fewer than 1014 operations per second. But because it remains unclear how much parallelism brains have, I use this larger estimate to be conservative. Theoretically, then, a perfectly efficient one-liter nanologic computer would provide the equivalent of about 10,000 times 10 billion human beings (or about 100 trillion human beings) in terms of brain capability. To be clear, I’m not arguing that this is achievable in practice. The point is that nanoscale engineering offers staggering amounts of headroom for future progress. Even a tiny fraction of a percent of theoretical maximums would be an utterly revolutionary new paradigm for computing—and allow nanoscale machines to attain useful amounts of computing power.

Many pathways to this type of mechanosynthesis and other nanotechnologies are currently being pursued.[80] Among them are DNA origami[81] and DNA nanorobotics,[82] bio-inspired molecular machines,[83] molecular Lego,[84] single-atom qubits for quantum computing,[85] electron beam–based atom placement,[86] hydrogen depassivation lithography,[87] and scanning tunneling microscope–based manufacturing.[88] There are several stealth projects within these spaces that are making steady progress, and—considering that superhuman engineering AI will be available by the end of the 2020s to solve remaining problems—we are on track for nanotechnology concepts using atom-by-atom placement to be implemented sometime in the 2030s.

In 2023 the value of physical products comes from many sources, especially raw materials, manufacturing labor, factory machine time, energy costs, and transportation. But convergent innovations will be dramatically reducing most of those costs in the coming decades. Raw materials will be cheaper to extract or synthesize with automation, robotics will replace expensive human labor, high-priced factory machines will themselves become cheaper, energy prices will fall due to better solar photovoltaics and energy storage (and eventually fusion), and autonomous electric vehicles will drive down transportation costs. As all these components of value become less expensive, the proportional value of the information contained in products will increase. Indeed, we are already going in that direction, as the “information content” of most products is rapidly increasing—and will ultimately get very close to 100 percent of their value. In many cases, this will make products cheap enough that they can be free to consumers. Again, we can look to the digital economy to see how this has already played out. As discussed in chapter 5, platforms like Google and Facebook spend billions of dollars on their infrastructure, but the average cost per search or per Like is so low that it makes more sense to make them totally free for users—with other revenue sources like ads being used to make money. In a similar way, it’s possible to imagine a future where people watch political ads or share personal...

data in order to get free nano-manufactured products. Governments might also offer such products as incentives for volunteer service, continuing education, or maintaining healthy habits." Kurzweil drops the facade and goes full Black Mirror, hahaha.

It will be a challenge to guarantee that these benefits are shared widely and fairly. That said, I am optimistic. The idea that wealthy elites would simply hoard this new abundance is grounded in a misunderstanding. When goods are truly abundant, hoarding them is pointless. Nobody bottles up air for themselves, because it’s easy to get and there’s enough for everyone.

Not breathing polluted air is a highly valued commodity, enough that the US dismantled its manufacturing for it.

While nanotechnology will allow the alleviation of many kinds of physical scarcity, economic scarcity is also partly driven by culture—especially when it comes to luxury goods. For example, to the naked eye, artificial diamonds are already indistinguishable from natural diamonds, but they sell for about 30 to 40 percent less.[91] This component of the price has nothing to do with the ornamental beauty of the diamonds, but rather with our cultural conventions that assign more value to diamonds that formed naturally. Likewise, paintings by the old masters aren’t really any better at sprucing up a living room than high-quality reproductions, but because people value their status as originals, they may sell for roughly a million times more.[92] So the nanotech manufacturing revolution won’t eliminate all economic scarcity. Historic diamonds and Rembrandts will remain scarce. But over the scale of generations, cultural values do change. Who’s to say whether the people who are now children will choose different values when they are adults? Or their own children?

In the 2020s we are starting the second phase of life extension, which is the merger of biotechnology with AI. This will involve developing and testing breakthrough treatments in digital biology simulators. Early stages of this have already begun, and with these techniques we will be able to discover very powerful new therapies in days rather than years. The 2030s will usher in the third phase of life extension, which will be to use nanotechnology to overcome the limitations of our biological organs altogether. As we enter this phase, we’ll greatly extend our lives, allowing people to greatly transcend the normal human limit of 120 years.[94]

At around age 110, the bodies of the oldest people start breaking down in ways that are qualitatively different from the aging of younger senior citizens.[97] Supercentenarian (110-plus) aging is not simply a continuation or worsening of the same kinds of statistical risks of late adulthood. While people at that age also have an annual risk from ordinary diseases (although the worsening of these risks may decelerate in the very old), they additionally face new challenges like kidney failure and respiratory failure. These often seem to happen spontaneously—not as a result of lifestyle factors or any disease onset. The body apparently just starts breaking down.

As de Grey explains, aging is like the wear on the engine of an automobile—it is damage that accumulates as a result of the system’s normal operation. In the human body’s case, that damage largely comes from a combination of cellular metabolism (using energy to stay alive) and cellular reproduction (mechanisms for self-replication). Metabolism creates waste in and around cells and damages structures through oxidation (much like the rusting of a car!). When we’re young, our bodies are able to remove this waste and repair the damage efficiently. But as we get older, most of our cells reproduce over and over, and errors accumulate. Eventually the damage starts piling up faster than the body can fix it. In a person in their seventies, eighties, or nineties, this damage will likely cause one fatal problem a significant amount of time before it causes several. So if science develops a drug to successfully treat an otherwise fatal cancer in an eighty-year-old, that person might expect to live almost another decade before something else kills him. But eventually everything starts failing at once and it is no longer effective to treat the symptoms of the damage caused by aging. Instead, longevity researchers argue, the only solution is to cure aging itself.

If you can live long enough for anti-aging research to start adding at least one year to your remaining life expectancy annually, that will buy enough time for nanomedicine to cure any remaining facets of aging. This is longevity escape velocity.[100] This is why there is sound logic behind Aubrey de Grey’s sensational declaration that the first person to live to 1,000 years has likely already been born. If the nanotechnology of 2050 solves enough issues of aging for 100-year-olds to start living to 150, we’ll then have until 2100 to solve whatever new problems may crop up at that age. With AI playing a key role in research by then, progress during that time will be exponential. So even though these projections are admittedly startling, and even sound absurd to our intuitive linear thinking—we have solid reasons to see this as a likely future.

In 1924 life expectancy in the United States averaged about 58.5 years, so babies born that year were statistically expected to die in 1982.[101] But medicine saw so many improvements during that interval that many of these individuals lived into the 2000s or 2010s.

Willfully misrepresents infant mortality decline as lifespan extension.

As I see it, the long-term goal is medical nanorobots. These will be made from diamondoid parts with onboard sensors, manipulators, computers, communicators, and possibly power supplies.[102] It is intuitive to imagine nanobots as tiny metal robotic submarines chugging through the bloodstream, but physics at the nanoscale requires a substantially different approach. At this scale, water is a powerful solvent, and oxidant molecules are highly reactive, so strong materials like diamondoid will be needed. And whereas macro-scale submarines can smoothly propel themselves through liquids, for nanoscale objects, fluid dynamics are dominated by sticky frictional forces.[103] Imagine trying to swim through peanut butter! So nanobots will need to harness different principles of propulsion. Likewise, nanobots probably won’t be able to store enough onboard energy or computing power to accomplish all their tasks independently, so they will need to be designed to draw energy from their surroundings and either obey outside control signals or collaborate with one another to do computation.

But nanobots won’t be limited to preserving the body’s normal function. They could also be used to adjust concentrations of various substances in our blood to levels more optimal than what would normally occur in the body. Hormones could be tweaked to give us more energy and focus, or speed up the body’s natural healing and repair. If optimizing hormones could make our sleep more efficient,[112] that would in effect be “backdoor life extension.” If you just go from needing eight hours of sleep a night to seven hours, that adds as much waking existence to the average life as five more years of life span!

As AI gains greater ability to understand human biology, it will be possible to send nanobots to address problems at the cellular level long before they would be detectable by today’s doctors. In many cases this will allow prevention of conditions that remain unexplained in 2023. Today, for example, about 25 percent of ischemic strokes are “cryptogenic”—they have no detectable cause.[123] But we know they must happen for some reason. Nanobots patrolling the bloodstream could detect small plaques or structural defects at risk of creating stroke-causing clots, break up forming clots, or raise the alarm if a stroke is silently unfolding.

Biological systems are limited in strength and speed because they must be constructed from protein. Although these proteins are three-dimensional, they have to be folded from a one-dimensional string of amino acids.[124] Engineered nanomaterials won’t have this limitation. Nanobots built from diamondoid gears and rotors would be thousands of times faster and stronger than biological materials, and designed from scratch to perform optimally.[125]

Nanomachines, son!

Thanks to these advantages, even our blood supply may be replaced by nanobots. A design by founding Singularity University nanotechnology cochair Robert A. Freitas called the respirocyte is an artificial red blood cell.[126] According to Freitas’s calculations, someone with respirocytes in his bloodstream could hold his breath for about four hours.[127]

Recall my estimate that the computation inside the human brain (at the level of neurons) is on the order of 1014 per second. As of 2023, $1,000 of computing power could perform up to 130 trillion computations per second.[130] Based on the 2000–2023 trend, by 2053 about $1,000 of computing power (in 2023 dollars) will be enough to perform around 7 million times as many computations per second as the unenhanced human brain.[131] If it turns out, as I suspect, that only a fraction of the brain’s neurons are necessary to digitize the conscious mind (e.g., if we don’t have to simulate the actions of many cells that govern the body’s other organs), this point could be reached several years sooner.

In the 2040s and 2050s, we will rebuild our bodies and brains to go vastly beyond what our biology is capable of, including their backup and survival. As nanotechnology takes off, we will be able to produce an optimized body at will: we’ll be able to run much faster and longer, swim and breathe under the ocean like fish, and even give ourselves working wings if we want them. We will think millions of times faster, but most importantly, we will not be dependent on the survival of any of our bodies for our selves to survive.

Guess I have a life goal to work towards.

Malevolent plagues like the Black Death arose from a combination of fast spread and severe mortality—killing about one third of Europe’s population[22] and reducing the world population from around 450 million to about 350 million by the end of the fourteenth century.[23]

It took thirteen years after its discovery to sequence full-length genome of HIV in 1996, and only thirty-one days to sequence the SARS virus in 2003, and we can now sequence many biological viruses in a single day.[31] A rapid response system would entail capturing a new virus, sequencing it in about a day, and then quickly designing medical countermeasures.

In May 2020 I wrote an article for Wired arguing that we should leverage artificial intelligence in order to create vaccines—for example, against the SARS-CoV-2 virus that causes COVID-19.[34] As it turned out, that is exactly how successful vaccines like Moderna’s were created in record time. The company used a wide range of advanced AI tools to design and optimize mRNA sequences, as well as to speed up the manufacturing and testing process.[35] Thus, within sixty-five days of receiving the virus’s genetic sequence, Moderna dosed the first human subject with its vaccine—and received FDA emergency authorization just 277 days after that.[36] This is stunning progress, considering that before COVID-19 the fastest anyone had ever created a vaccine was about four years.[37]

(For a rough sense of how much it would cost rogue actors to build nuclear weapons, consider the example of North Korea, a pariah state denied most access to outside assistance. The South Korean government estimates that the North’s nuclear weapons program cost between $1.1 billion and $3.2 billion in 2016, the year it successfully developed nuclear-armed missiles.)[39] By contrast, biological weapons can be very cheap. According to a 1996 NATO report, such weapons could be developed for $100,000 (around $190,000 in 2023 money) by a team of just five biologists in the space of a few weeks, without any exotic equipment.[40] When it comes to impact, a 1969 expert panel reported to the United Nations that biological weapons were about eight hundred times as cost-effective as nuclear weapons at targeting civilians—and biotech advances in the five decades since have almost surely increased that ratio significantly.[41]

Nano-based weapons could include tiny drones that deliver poisons to targets without being detected, nanobots that enter the body in water or as an aerosol and tear it apart from within, or systems that selectively target certain groups of people of any description.[42] As nanotechnology pioneer Eric Drexler wrote in 1986, “ ‘Plants’ with ‘leaves’ no more efficient than today’s solar cells could out-compete real plants, crowding the biosphere with an inedible foliage. Tough, omnivorous ‘bacteria’ could out-compete real bacteria: they could spread like blowing pollen, replicate swiftly, and reduce the biosphere to dust in a matter of days. Dangerous replicators could easily be too tough, small, and rapidly spreading to stop—at least if we made no preparation. We have trouble enough controlling viruses and fruit flies.”[43]

The most commonly discussed worst-case scenario is the potential creation of “gray goo”—self-replicating machines that consume carbon-based matter and turn it into more self-replicating machines.[44] Such a process could lead to a runaway chain reaction, potentially converting the entire biomass of the earth to such machines. Let’s consider how long it might take for the earth’s entire biomass to be destroyed. The available biomass has on the order of 1040 carbon atoms.[45] The carbon atoms within a single replicating nanobot might be on the order of 107.[46] The nanobot therefore would need to create 1033 copies of itself—not all directly, I should stress, but via iterated replication. The nanobots in each “generation” might just create two copies of themselves, or another small number. The astoundingly large numbers come from this process being repeated again and again with the copies and the copies of the copies. So that’s about 110 generations of the nanobots (since 2110 = 1033), or 109 if the nanobots in previous generations stay active.[47] Nanotechnology expert Robert Freitas estimates a replication time of about one hundred seconds, so under ideal conditions the gray goo wipeout time for that much carbon would be around three hours.[48] However, the actual rate of destruction would be much slower, because the world’s biomass is not laid out in a continuous block. The limiting factor would be the actual movement of the front of destruction. Nanobots cannot travel ve...

When people are presented with the prospect of radical life extension, two objections are quickly raised. The first is the probability of running out of material resources to support an expanding biological population. We frequently hear that we are running out of energy, clean water, housing, land, and the other resources we need to support a growing population, and that this problem will only be exacerbated when the death rate starts to plummet. But as I articulated in chapter 4, as we begin to optimize our use of the earth’s resources, we’ll find they are thousands of times greater than we require. For example, we have nearly ten thousand times the sunlight we need to theoretically meet all of our current energy needs.[83] The second objection to radical life extension is that we will become profoundly bored doing the same things over and over again for centuries. But in the 2020s we will have virtual and augmented reality delivered in very compact external devices, and in the 2030s we will have VR and AR connected directly to our nervous systems by nanobots feeding signals to our senses. We will thereby have radical life expansion in addition to radical life extension. We will inhabit vast virtual and augmented realities limited only by our imagination—which itself will be expanded. Even if we lived hundreds of years, we would not exhaust all the knowledge there is to gain, and all the culture there is to consume. AI is the pivotal technology that will allow us to meet...

Cassandra: So you anticipate a neural net with sufficient processing power to be able to exceed all capabilities by humans by 2029. Ray: Correct. They are already doing that with one capability after another. Cassandra: And when they do that, they will be far better than any human in every skill possessed by any human. Ray: Correct. In one area after another, they will be better than all humans by 2029.

Cassandra: And you are also expecting that by the early 2030s we will have a means of going inside the brain and connecting to the top levels of the neocortex, both to tell what is going on and to activate connections. Ray: Right.

Ray: By the middle of the 2020s we will have a means of interacting with a computer that is thousands of times faster than keyboarding: fully immersive virtual reality with full-screen video and audio. We will see and hear ordinary reality, but it will be interlaced with two-way communication with our computers. That is almost as fast as a connection with the top layers of our neocortex. This will ultimately replace interaction by keyboards.

Cassandra: I agree, but extending the neocortex into the cloud is a very different kind of progress than better external brain extenders. Ray: Yes, your point is valid, but I do think we will achieve an extension of the neocortex by the early 2030s. So the in-between period probably won’t be very long.

Cassandra: You’ve argued, though, that within a few years our brains will effectively be cloud extensions. Ray: We’re actually already doing that. And whatever philosophical significance you see for our biological brain, we are not taking that away, either. Cassandra: But the biological brain at that point will be pretty insignificant. Ray: But it is still there, and it will retain all its fundamental qualities.