This piece originally appeared on TomDispatch.

My wife and I recently drove in Virginia and were once again surprised that the GPS technology on our mobile phones can guide us through dense highways, bypass road accidents, and reach our precise destination. The artificial intelligence (AI) behind the soothing sounds that tell us where to turn has replaced passenger seat navigators, maps, and even traffic information updates on the radio. How did we survive before this technology entered our lives? Of course, we survived, but sometimes we do lose our way.

My reverie was interrupted by the toll booth. It is empty, as are all other stalls in this special toll plaza. Most cars pass quickly with an EZ pass because one automated device communicates seamlessly with another. Unfortunately, our rental car does not.

So I am going to pay by credit card, but the booth does not have a credit card reader.

Well, I think, when I take out my wallet, I will pay $3.25 in cash.

As it happens, the stall only took coins. Who was driving around in his or her pocket?

I wanted to ask someone this question, but of course I was surrounded by silent machines. So, I just drove through the electronic railings to prepare for the bills that would be received once the plaza's automated system photographed and tracked our license plates.

In a completely mundane way, I have just experienced the ancient conflict between the restrictive and liberating nature of technology. The arrow that can provide you with dinner food may end up stuck in your own skull. The car that takes you to the beach on holiday will cause the tide to rise-through carbon emissions and increased temperature-that one day may wash away the gems of this very coastal place. The laptop that connects you to the online world can also serve as a channel for hackers to steal your identity and clear your bank account.

In the last century, technology has reached a real watershed moment, and humans have also gained the ability to destroy the entire planet by using the power of atoms. Now, thanks to artificial intelligence, technology is pushing us to a new inflection point.

Science fiction writers and technologists have long feared that robots will take over the future of the planet after gaining perception. Creating a machine with human-like intelligence might one day deceive us into believing that it is one of us. This is often described as a "singularity" and it is chilling. Respected scientists like Stephen Hawking believe that such a singularity will actually mark the "end of mankind."

This will not be an impossible remote event like the sun exploding in a supernova in billions of years. According to a public opinion survey, artificial intelligence researchers estimate that by 2050, the probability of a singularity occurring is at least 50-50. In other words, if pessimists like Hawking are right, then robots are likely to send humans before the climate crisis occurs.

Neither the artificial intelligence that powers GPS nor the artificial intelligence that controls that frustrating toll station has reached human level intelligence—not even close. But in many ways, this stupid robot has already taken over the world. Automation is currently replacing millions of workers, including those former toll gate operators. "Smart" machines such as drones have become an indispensable part of the war. More and more artificial intelligence systems are deployed to monitor our every move on the Internet, through mobile phones, and when we venture into public places. Algorithms are replacing teaching assistants in the classroom and affecting sentencing in courts. Some of the loneliest of us have begun to rely on robot pets.

With the continuous improvement of artificial intelligence capabilities, the inevitable political question will become: To what extent can these technologies be contained and regulated? Yes, nuclear elves, like other technologies (biology and chemistry), can cause large-scale destruction on Earth that was previously unimaginable. However, with artificial intelligence, the day of the singularity is still in the future, even if it is fast approaching. At least in theory, it is still possible to control such results before doing nothing, but afterwards, you can only play a non-proliferation game of whack-a-mole.

As long as humans continue to behave badly on a global scale-wars, genocide, carbon emissions that threaten the planet-it is difficult to imagine that anything we create, no matter how smart, will behave in different ways. However, we still dream of a certain god in the machine, the god in the machine, can appear like magic and save us from our own hands.

In the early 1940s, science fiction writer Isaac Asimov formulated his famous Three Laws of Robots: Robots must not directly or indirectly harm humans; they must obey our orders (unless doing so violates the first law) ); And they must protect their existence (unless self-protection violates the first two rules).

Many writers have tried to update Asimov. The latest is the legal scholar Frank Pasquale, who designed four laws to replace Asimov's three laws. Since he is a lawyer rather than a futurist, Pasquale is more concerned with controlling today's robots than assuming tomorrow's machines. He believes that robots and artificial intelligence should help professionals, not replace them; they should not impersonate humans; they should never be part of any arms race; and their creators, controllers, and owners should always be transparent.

However, Pasquale's "laws" run counter to our current artificial intelligence trends. The prevailing AI spirit reflects what can be considered as the main guiding principle of Silicon Valley: Act fast and break the norm. Most importantly, this subversive concept requires technology to continuously reduce labor costs and often makes itself obsolete.

In the global economy, artificial intelligence has indeed helped certain professionals — such as Facebook’s Mark Zuckerberg and Amazon’s Jeff Bezos, who happen to be one of the richest people on the planet — but it is also Replace millions of us. In the military field, in the upcoming world of robot warfare, automation is pushing boots off the ground and looking to the sky. Whether it's Siri, a robot that guides increasingly frustrated callers through an automated phone tree, or an artificial intelligence that checks Facebook posts, the purpose is to fake human beings—"machines like me", as Ian McEwan said in The title as in his 2019 novel-at the same time, it hides the string that connects the creation with its creator.

Pasquale hopes to brake trains that not only leave the station but are no longer under the control of the engine driver. It is not difficult to imagine where this out-of-control phenomenon will end. Technological pessimists have adopted an abnormal approach when describing the resulting disaster. For example, Nick Bostrom wrote about a sandstorm of self-replicating nano-robots that suffocated all living things on the planet-the so-called gray slime problem-and a way through "hijacking the political process." "Artificial intelligence that seizes power.

Since they are only interested in self-protection and copying, rather than protecting humans or obeying their orders, this sentient machine will obviously tear Asimov's rulebook. Futurists have jumped into the breach. For example, Ray Kurzweil predicted in his 2005 book "The Singularity Is Near" that robots will gain consciousness around 2045, and he proposed "to prohibit self-replicating physical entities , These entities contain their own self-replicating code". Elon Musk is another billionaire industrialist who is not the enemy of innovation. He called artificial intelligence the "biggest existential threat" to mankind and supported the prohibition of future killer robots.

In order to prevent various worst-case scenarios, the EU proposes to control artificial intelligence according to the degree of risk. Some products belonging to the EU's "high risk" category must obtain a Good Housekeeping approval seal (Conformité Européenne). On the other hand, artificial intelligence systems "are considered to pose a clear threat to people's safety, livelihoods, and rights" and will be completely banned. For example, such obvious dangers include biometric recognition of personal data through facial recognition and other methods, and the version of China's social credit system that artificial intelligence helps track individuals and assess their overall credibility.

Technological optimists can predictably criticize what they believe to be over-expansion in Europe. They believe that this kind of control over artificial intelligence will hinder research and development, and if the United States follows suit, China will gain insurmountable technological advantages in this field. Entrepreneur Sid Mohasseb wrote in Newsweek: “If EU member states-and their allies on the other side of the Atlantic-seriously compete with China and maintain their position of power (and the quality of life they provide to citizens)," "They need to call for these regulations to be redrafted. Growth and competition are considered at least as important as regulation and safety."

However, Mohasseb's concerns are misleading. In fact, the regulator he is very worried about is now playing a catch-up game. In the economy and the battlefield, artificial intelligence is already indispensable in the two fields of human activity alone.

The ongoing Covid-19 pandemic has exposed the fragility of global supply chains. The world economy has almost stagnated in 2020, one of the main reasons: the health of human workers. The spread of infection, the risk of infection, and efforts to contain the pandemic have all caused workers to disappear from the workforce, sometimes temporarily, sometimes permanently. The factory closed, the transportation network gap widened, and the store's business was taken away by online sellers.

The desire to reduce labor costs is a major factor in product prices, prompting companies to look for cheaper workers overseas. For these cost-cutters, the prospect of eliminating workers altogether is even more tempting. Long before the pandemic hit, companies had already begun to turn to automation. By 2030, as many as 45 million American workers will be replaced by robots. The World Bank estimates that they will eventually replace 85% of the staggering jobs in Ethiopia, 77% in China and 72% in Thailand. "

The pandemic has not only accelerated this trend, it has also exacerbated economic inequality, because at least for now, robots tend to replace the least skilled workers. In a survey conducted by the World Economic Forum, 43% of companies stated that they would reduce their labor force by using more technology. "Since the outbreak of the pandemic," NBC News reported,

“Food manufacturers have increased automation, allowing facilities to maintain social distancing while maintaining output. The factory digitizes controls on machines so that workers working at home or elsewhere can operate them remotely. New devices that can flag or predict failures have been installed. Sensors, allowing the team of inspectors operating on schedule to be reduced to on-demand maintenance personnel."

In an ideal world, robots and artificial intelligence will increasingly take on all dirty, dangerous, and degrading jobs on a global scale, allowing humans to do more interesting jobs. However, in the real world, automation often makes jobs dirtier and more dangerous, for example, by speeding up the work of surplus labor. At the same time, robots began to encroach on more interesting work that is generally considered to be done by architects and product designers.

In some cases, artificial intelligence has even replaced managers. Stephen Normandin, an Amazon contract driver, discovered that the artificial intelligence system that monitors his efficiency as a deliveryman will also use automated emails to fire him when it determines that he does not meet the requirements. Jeff Bezos may resign as CEO of Amazon, but the robot is rapidly climbing its corporate ladder and may prove to be at least as ruthless as he was before, if not more.

It may be particularly difficult to mobilize against such a robot as an alternative to an army, because corporate executives are not the only ones who come up with welcome mats. Since fully automated manufacturing in the "dark factory" does not require lighting, heating, or labor to drive to the site, such production can reduce a country's carbon footprint-a potential enticing factor for advocates and advocates of "green growth" . Politicians are eager to achieve their Paris climate goals.

Conscious robots may not need to devise clever strategies to take over the world. Humans may be very willing to hand over the keys to the kingdom to semi-intelligent machines.

The new fog of war

Facts have proved that the war between Armenia and Azerbaijan in 2020 is different from any previous military conflict. Since the 1980s, the two countries have been competing for the disputed mountain enclave Nagorno-Karabakh. After the collapse of the Soviet Union, Armenia proved a clear victor in the conflicts in the early 1990s, occupying not only disputed territories but also parts of Azerbaijan.

In September 2020, as tensions between the two countries intensified, Armenia prepared to defend these occupied territories with a well-equipped army of tanks and artillery. However, due to its fossil fuel exports, Azerbaijan’s expenditure on the most modern military preparations far exceeds that of Armenia. Despite this, Armenian leaders often tout their army as the best in the region. In fact, according to the 2020 Global Militarization Index, the country's militarization level is second only to Israel. 

However, Azerbaijan was the decisive victor of the 2020 conflict, regaining Nagorno-Karabka. Reason: automation.

"Azerbaijan used drone fleets purchased from Israel and Turkey to track and destroy Armenian weapon systems in Nagorno-Karabakh, smash its defenses and advance quickly," Robin Dixon of the Washington Post reported. “Armenia discovered that Nagorno-Karabakh’s air defense systems, many of which were older Soviet systems, were unable to withstand drone attacks, and losses increased rapidly.”

Armenian soldiers, known for their ferocity, are often frightened by the semi-autonomous weapons above them. "Soldiers on the ground know that they may be hit by drones hovering overhead at any time," noted Mark Sullivan in the business magazine Fast Company. "The drones are very quiet and they didn't hear the hum of the propellers until it was too late. Even if the Armenians did shoot down one of the drones, what did they really accomplish? They just destroyed a piece and will be replaced. Machine." 

In the war on terrorism in Afghanistan, Iraq, Pakistan, Somalia, and other parts of the greater Middle East and Africa, the United States took the lead in using drones against various non-state opponents. But in the 2020 election campaign, Azerbaijan is using this technology to defeat a modern army. Now, every military will have to not only integrate increasingly powerful artificial intelligence into its offensive capabilities, but also to defend against new technologies.

In order to stay ahead in this field, it is foreseeable that the United States is investing a lot of money in the latest technology. The Pentagon’s new budget includes the “largest ever” R&D requirement, which includes an AI down payment of nearly $1 billion. As TomDispatchregular Michael Klare wrote, the Pentagon even learned a lesson from the business world and began to replace its war manager-the general-with a huge network of interconnected automated systems, known as Joint Global Command and Control (JADC2 ).

Any such transfer of greater responsibility to the machine will result in what the mathematician Kathy O’Neill calls a "mathematical destructive weapon". In the global economy, artificial intelligence has replaced humans in the upstream and downstream of the production chain. In the world of war, whether it is due to human design or computer errors, artificial intelligence may eventually wipe out mankind.

After all, during the Cold War, only the last-minute intervention of individuals on both sides could ensure that nuclear "missile attacks" detected by the Soviet and American computers-which turned out to be birds, abnormal weather or computer failures-would not suddenly happen. Nuclear war. Removing humans from the chain of command, machines can carry out such genocide on their own.

Dear reader, the error is not in our robots, but in ourselves.

In my new novel Songlands, mankind faces a series of terrible choices in 2052. Having failed to control carbon emissions for decades, the world is at a point where it cannot turn back, and it is too late to fix traditional policies. The only thing left is the scientific Wan Fu Mary Pass, which is a geoengineering experiment that may fail, or worse, have terrible and unexpected consequences. The AI ​​responsible for ensuring the success of the experiment may or may not be trustworthy. My dystopia, like many others, is actually about the narrowing of options and the reduction of hope. This is our current development trajectory.

However, we still have options. We can fundamentally turn to clean energy and use resources for the entire world (not just its richer parts) to achieve leapfrog development together. We can implement reasonable supervision of artificial intelligence. We can discuss the details of such plans in democratic societies and participatory multilateral forums.

Or, we can wait for the post-Trump saviour to save us by raising our hands because of political differences that we cannot bridge. Technological optimists hope that automation will free us and save the planet. Laissez-faire enthusiasts continue to believe that the invisible hand of the market will mysteriously lead capital to innovations that save the planet, rather than SUVs and plastic trinkets.

These are hallucinations. As I wrote in Songlands, we always hope that someone or something will save us: "God, dictator, technology. For better or worse, the only response to our call for help is an echo."

In the end, robots can't save us. This is a job that cannot be outsourced or automated. This is work that only we can accomplish.

To stay up to date with such important articles, please register here to receive the latest updates from TomDispatch.com.

John Feffer is the author of the new dystopian novel "Divided Lands" and is the director of the Foreign Policy Focus of the Policy Research Institute.

Copyright © 2021 Salon.com, LLC. It is strictly forbidden to copy materials from any salon page without written permission. SALON ® is registered as a trademark of Salon.com, LLC in the United States Patent and Trademark Office. Associated Press Article: Copyright © 2016 The Associated Press. all rights reserved. Do not publish, broadcast, adapt or redistribute this material.