Elliott Earnshaw reflects on the future of humanity and the challenges we will face in the future.
We walk along a path cut into the edge of a cliff. It goes up. But as the path goes up it narrows. With each step, the road behind us crumbles. There is no turning back. There are only two possibilities; we make it to the top, or we plunge to our deaths. This is the new normal.
Australian philosopher Toby Ord has coined the phrase “the precipice” to describe our current age of existential risk. His new book investigates the probability of humanity’s survival and how we can safeguard our future. This article draws from Ord’s ideas and takes a look at the existential risks we face and how we can use technology to “make it to the top”.
Thus far in human history, we have stumbled many times but walked on relatively unscathed. Technological advancements have made our lives better than any of our ancestors could have imagined. Globally, life expectancy at birth has increased from roughly 29 years for most of human history, to 73 years in 2019. As recently as 1800, no country in the world had a life expectancy above 40. Less than 10 percent of people today live in extreme poverty, compared to over 90 per cent two centuries ago. The global child mortality rate is more than 10 times lower than in the past. There is good reason to believe that technology can continue to solve most, if not all, major problems we face as a species.
However, technology can also create even greater problems. Problems such as nuclear war, engineered pandemics, environmental catastrophe, climate change and unaligned artificial intelligence. As American sociobiologist E.O. Wilson said, ‘we have palaeolithic emotions, medieval institutions and god-like technologies.’ This has never been truer than it is today. It is these god-like technologies which pose an existential risk to humanity. We are the makers of our own destiny, and potentially the makers of our own demise. We owe ourselves a duty to be aware of these risks and educate others so that collectively we may diminish them.
Perhaps the most obvious existential threat comes from nuclear war. Some fifteen thousand nuclear weapons are scattered around the world in the hands of 9 often opposing states. Whilst the claim that we have enough nuclear weapons to destroy the world many times over is inaccurate; we could certainly destroy the world as we know it. And there have been numerous times in history where we have come close. According to William J. Perry, former US Secretary of Defence and nuclear weapons expert, the main threat comes from false alarms in times of heightened tension.
For example, at midnight on 25 October 1962, a guard at Deluth Airbase spotted a figure trying to climb the fence into the base. Believing this may be an attempt of sabotage preceding a nuclear strike, the guard activated the sabotage alarm but hit the wrong switch and instead set off an alarm which signified an incoming nuclear attack. This automatically set off similar alarms at other bases. In an instant, nuclear-armed F-106A’s were ordered into the air “fully believing that a nuclear war was starting”. As it turned out, the intruder was a bear. Fortunately, the error was realised and the planes stood down. Several similar close calls occurred all throughout the cold war.
Control of nuclear weapons is also scarily unconstrained. Perry explains in his 2020 book ‘The Button’ that currently, the US President has unrestrained authority to launch nuclear weapons. The US does not have a ‘no first strike policy’. Meaning they can launch a nuclear attack without any threat from another nation in a completely non-retaliatory context. It takes less than ten minutes to do so and once launched, nuclear-armed ballistic missiles cannot be recalled.
Both the US and Russia continue to invest trillions in new nuclear weapons and false alarms and cyberattacks presenting false warnings of attack are very real possibilities. The cumulative result of these factors is a trigger-happy system that makes it far too likely to accidentally stumble our way into nuclear war.
And the threat only seems to be getting worse over time. Technological progress is making nuclear weapons more accessible for more entities, increasing the risk of nuclear technology entering the hands of ideological and omnicidal groups. Nuclear terrorism is one of the reasons Perry believes that ‘the likelihood of a nuclear catastrophe is greater today than during the cold war, and the public is completely unaware of the danger.’
This sentiment is reflected by the doomsday clock which is now at 100 seconds to midnight; the closest it has been since being founded in 1947. This represents the closest we have ever been to the end of the world. The world’s experts agree that our existence is at its most vulnerable point in human history. Put simply, we have developed the power to destroy ourselves without the wisdom to ensure we don’t.
Biotechnology poses another major threat to humanity. Scientists can already create and genetically modify pathogens in labs. It is already possible to engineer pathogens to be highly contagious, lethal, and largely undetectable. It is easy to imagine a pathogen which has a 100 percent fatality rate, is transmittable through the air, can survive outside of a host for 2 months and cannot be detected until death. It would only take one such ‘designer virus’ to, accidentally or otherwise, escape the lab to drastically disrupt world order.
The high population density, interconnectedness and globalisation of the modern world further contribute to the potency of a pandemic. It could quickly take hold, potentially killing billions of people and destroying civilisations which have taken thousands of years to build. COVID-19 has caused immense global disruption yet is relatively harmless compared with many such genetically modified viruses.
As with nuclear weapons, biotechnology and information about pathogens are becoming progressively more accessible to backyard scientists. For example, the entire DNA sequence of polio is available for download off the internet. As more people gain access to techniques of gene editing, the chance of an engineered pathogen being used with omnicidal intent increases. Certain sensitive information pertaining to biological weapons could be leaked from the lab. In the wrong hands, this could be used for biological terrorism and have truly devastating consequences.
Likewise, anthropogenic climate change poses a significant threat to humanity. Carbon dioxide levels in the atmosphere have risen from about 280 parts per million prior to the industrial revolution to 412 in 2019, the Earth’s climate has warmed 1 Degree Celsius, and sea levels have risen 23 centimetres. The future consequences of anthropogenic climate change are expected to be existentially significant. Whilst climate change won’t kill us all, it may contribute to something which does.
Artificial intelligence has the potential to revolutionise the world in the most spectacularly positive ways. As machines begin to think for themselves, form their own autonomy and gain more control of virtual and physical tasks, we will be able to solve countless problems in ways we never thought possible. When 300 top researchers in machine learning were asked in 2016 when an AI system would be ‘able to accomplish every task better than human workers’, on average they estimated a 50 percent chance of this happening by 2061. Such a system is referred to as strong AI or artificial general intelligence (AGI).
However, AGI without objectives clearly aligned with those of the human population may develop its own ambitions. It may look to seize control of the world. There are concerns about AI entrenching social discrimination, producing mass unemployment, supporting oppressive surveillance, or even disobeying human commands and redirecting its own reward function toward the ultimate goal of its survival, at the expense of ours. Physical robots are not necessary for this either. Absolute control can be achieved through manipulation using words and ideas, all hidden from us until it is too late. A generally intelligent machine could make us humans work for it without us ever realising.
Oxford philosopher Nick Bostrom’s vulnerable world hypothesis compares human creativity to pulling balls out of a giant urn of inventions, each ball representing a possible technological discovery. Some balls are lightly coloured; they pose little harm: the invention of the mobile phone, the electric car or Wi-Fi. Some balls are darker; agent orange or landmines come to mind. Then there are black balls. To pull out a black ball would spell the end of humanity. Fortunately, we are yet to find one.
But often when we reach into the urn, we do not know what coloured ball we will grasp. And when we take a ball, we cannot put it back in the urn. Technology never retreats. This is to say that we face existential risks from things which we have not even yet contemplated. The scope of such threats increases the further we travel up the path along the cliff.
We exist in a precarious position. With an ever-expanding array of existential risks, the future may look bleak. But technology also provides the possibility to improve our lives in ways previously unimagined. We have the ability to overcome the challenges we face. If we play our cards right, we can make our way to the top and avoid falling. Our actions determine the fate of the trillions of people to come, whose very existence lies in the balance.
We need to proceed with caution and educate people about the risks we face. We must unite and use our collective wisdom to solve the problems which we have created. The new normal is one of immense technological opportunity but it is also one of grave danger.
Adapted from an essay entered in the Professor Ian Jack Essay Prize.
Image: Elliott Earnshaw
