Looking for inspiration recently I found Edge.org, a website that purports to “seek out the most complex and sophisticated minds” and put them to work on the world’s most difficult problems. A bit presumptuous, but I was intrigued enough to dig a bit deeper. This year in 2013, the contributors (which range from psychologists to futurologists to hacker-musicians) have been set the task of describing those things we should be most worried about as a civilization. Cool idea, disappointing execution.
The problem with the majority of the responses is that everyone seemingly has a book to sell, so they have re-framed questions to fit their narrow and specialized areas of expertise. To a man with a hammer, everything looks like a nail; to an evolutionary scientist, everything looks like a problem of natural selection. The linguists think our biggest problem is language, the political scientists think our biggest problem is a breakdown of democracy, and so on.
Since I do not claim expertise in a particular field and since I don’t have a book to sell, I feel I can attack the problem from more epistemologically neutral ground. Here are four things I really think we should be worried about as a civilization. They are not doomsday scenarios: the usual asteroid impact, vacuum fluctuation, supervolcano and pandemic were already taken (and discussed ad nauseam in places like New Scientist). To invert the anthropic principle, a world in which humanity no longer exists is not interesting enough to discuss.
The four things I’ve chosen represent real, life-changing developments that may not wipe humanity from existence, but would seriously disrupt the continued and expected trajectory of our planet. And, importantly, any of them could happen tomorrow.
You can call them Black swans if you like, and they do meet some of the criteria of Taleb’s popular theory – they are high impact events that are not easy to compute with our current models of prediction. They are also ‘jumps’ in history, rather than smooth continuities from current events.
Figure 1: Four low-probability, high-impact events
Scenario 1: Contact with ET
We could look for analogies with the way that Copernican thought de-centered and de-privileged humanity’s place in the cosmos, but the truth is that we have no cognitive models to assess the impact of meeting a cosmological ‘other’. Empirical observation tells us that aliens do not exist (very few of us claim to have observed one). But on the other hand, a growing body of empirical evidence points to the near certainty that life exists elsewhere in the universe. So on one hand we have astronomy and the tools of detection growing more and more sophisticated, and on the other hand we have the deafening silence of Fermi’s paradox in observed local phenomena. For those of us who follow developments in cosmology, the tension sometimes feels almost unbearable.
The implications of contact, if it came, are hard to predict. Certainly it would give a much-needed boost to funding in science and space exploration. It could crucially ignite the imaginations of a younger generation of school kids who have grown up in a period of economic and cultural malaise and who yearn for a big metanarrative like my parents had in the 1960s with the cold war and I had in the 1980s with cybernetic technology. Reagan famously said on a number of occasions that discovery of an alien civilization would do more for world peace and shared purpose on this planet than any earth-bound effort could accomplish. Maybe he was right, but on the other hand a second Copernican revolution could intensify ideological schisms between scientific and non-scientific societies, creating a ‘reality gap’ much larger than the one we currently observe in politics.
We are certainly unprepared for alien contact, and that is another reason why it deserves to be listed here. With a potential for such high impact, we are woefully unprepared: ontologically, culturally, strategically. We should have our best minds working on ways to understand, model and prepare for the eventuality of contact with extraterrestrial life forms.
Scenario 2: Physics breakthrough
Physics has run up against a brick wall in being unable to unify the standard model with gravity, dark matter and dark energy. We don’t know if the universe is deterministic or probabilistic. We don’t know what preceded the big bang, or what will come at the end of observable time. While gaps in our knowledge are cause for frustration, they also represent an opportunity. The blank spots on the map in theoretical physics might contain particles and interactions that could be harvested in new ways to generate energy, to communicate or to manipulate the fabric of spacetime. Consider that one of the last major paradigm shifts in physics – the discovery of atomic nuclei – heralded the most destructive and disruptive period in human endeavor. A new physics breakthrough could do much the same. It might offer a solution to our current energy shortfalls, while at the same time introducing new risks and unforeseen problems.
The implications of a new physics paradigm for energy are enormous. Imagine that we discover a way to generate high-efficiency energy without producing large quantities of waste (the infamous ‘cold fusion’). The immediate effect would be a significant drop in fuel prices, making our economies more efficient and more productive. Cities would become cleaner and cheaper to run. We might develop new forms of thrust that could lift objects into orbit and interplanetary space at low cost. Suddenly private space exploration and colonization of the inner solar system would be in reach.
A new physics paradigm might bring with it a host of other related innovations that could have implications for fields like communication, computation, materials, optics, and medicine. If we think about everything that has been touched by the discovery of atomic energy, we begin to get a sense of what a major physics discovery could do. Whether such a breakthrough is possible, and what it might look like, are by nature unpredictable.
Scenario 3: Artificial intelligence
Although I find some posthumanist arguments compelling, I am skeptical about the claim that artificial intelligence will emerge automatically alongside accelerations in the complexity of computer hardware. That fact is that we simply don’t know how to create an artificial intelligence, or even whether it is possible. Yes we can build smaller and smaller transistors, ensuring that the exponential curve of Moore’s law will hold up in the foreseeable future, soon delivering us computers that have similar levels of complexity to a human brain (by the 2030s). But those computers will not be ‘alive’ unless we find some way to imbue them with the spark of conscious self-awareness.
The thing is that because we don’t have a reliable working theory of artificial intelligence, it could sneak up on us from anywhere. A video game designer has a reasonable chance of accidentally stumbling upon the secret (think Alexander Fleming). Or perhaps AI can emerge and self-assemble from dumb networks, the likes of which are proliferating across the planet unregulated. A secret military programme could ‘brute force’ the problem using mind bogglingly complex hardware.
A staggering number of computer scientists believe that artificial intelligence is not only possible but inevitable. A staggeringly small number of individuals have given thought to the civilizational risks such an event would present.
One person who has thought deeply about the existential risk represented by computer intelligence is Nick Bostrom. His central argument is that computers seem to be very good at optimization for problem solving, but not particularly good at nuance or understanding of the big picture (something that 2 million years of evolution on the plains of Africa has bestowed to humans). Combining a highly capable tactical thinking machine with single-minded objectives could be potentially dangerous. The bottom line is that in all of human history we have never had to contend with an intellect greater than our own. The impact could be as wonderful or as terrible as we can possibly imagine.
Scenario 4: Life extension
Experiments on mice, using a variety of therapies and techniques (calorie restriction, supplements, gene therapy) have produced promising results, with some individual techniques increasing lifespans by as much as 40%. If our current medicine can extend the lifespan of other mammals consistently using scientifically understood pathways, we are surely within the ballpark of a discovery that could do the same for humans. Our bodies are more complex, but a comparable 40% extension in human lifespan would correspond to an additional 28 years or more, pushing life expectancy well past 100 years for those alive today. Based on rapid advancements in biology, assisted by computer technology and our growing understanding of the human genome, I anticipate that such a breakthrough is possible within the next two decades.
And that may be only the beginning. Combining multiple techniques synergistically might extend human life expectancy by 50% or 60%, pushing us well past 120 years or more. These techniques might be possible without even having to drill too deeply into the human genome. Some of these effects could be achieved with chemical compounds ingested or injected into the body, subsequently producing beneficial effects either by ‘tricking’ or repairing damaged cells. At the same time that we push the survivability of the human body against the decay of ageing, we will also be making scientific progress in the fight against diseases like Alzheimer’s and cancer, pushing our survivability even further. We already possess promising therapies to fight both of those diseases, and we might be less than a decade away from more effective treatments.
I suspect that life extension will be one of the most contentious political issues of this century. The prospect of some humans living much longer than has ever been biologically possible is an ethical minefield: who should have access to these techniques, who will bear the economic burden of network externalities implied by an immortal population, how will we conceive of healthcare and human rights in a world on the verge of immortality?
Of the four scenarios outlined here, I actually think this one is most likely to occur. Many of the problems of ageing are already understood by science, at least in principle. The challenge is one of applied knowledge, something human beings have proved exceedingly good at.