Investor.bg
Last week, I traveled a considerable distance to attend an informal lecture on artificial intelligence (AI) given by the head of one of the largest intelligence agencies in the West. Like many people, even those familiar with the world of defense and security, I know next to nothing about AI and the revolution that will affect everyone’s lives, Max Hastings tells Bloomberg.
The trip turned out to be not worth the gas. The think tank audience was bombarded with half an hour of boring talk about how important AI will be, the need to regulate it and use it wisely. However, our speaker did not bother to explain what AI will do, beyond the obvious – think autonomously. I wondered as I left if they really knew what it was all about.
I’ll start with this story, so if you’re feeling as clueless as I am, rest assured, because you have company. But we can all understand the fact that AI will play a key role in decision-making of all kinds, commercial, governmental, and military. On the battlefield, almost all future weapons systems will have AI capabilities, such as synchronizing swarms of drones and responding to approaching missiles with literally superhuman speed. There are concerns, however, that even the most optimal AI moves will still encourage escalation.
Some experts go even further in their predictions. In a new book, completed and co-authored with tech gurus Eric Schmidt and Craig Mundy, the late Henry Kissinger argues that “immune to fear and preference, AI introduces a new possibility for objectivity in strategic decision-making.” Kissinger and his colleagues see the possibility that machines will behave more rationally, and therefore better, than aggressors like Russian President Vladimir Putin.
Most of us are more cautious and indeed nervous. If we didn’t trust that Chinese President Xi Jinping and US President-elect Donald Trump would make wise decisions about peace or war, we are even more terrified by the idea of machines having power over their judgments. Stephen Feldstein of the Carnegie Endowment for International Peace says that AI is “helping to accelerate a new era of geopolitics” that “will inevitably shape the functioning of democracy and governance.”
The increasing power of chatbots will enable malicious actors to significantly influence political opinion by publishing vast amounts of propaganda as fact, further poisoning public discourse in a way that many right-wing media outlets have already achieved through social media.
Given that we see airline computer systems crashing and government databases being hacked every day, it seems terrifying to think of a future in which a glitch in the system could launch a nuclear missile. We simply don’t believe the “experts” who seek to assure us that AI will not be given the power to do such a thing.
Many of us have seen Stanley Kubrick’s superb 1964 film satire Dr. Strangelove, in which a mad American general sends a group of strategic bombers against the Soviet Union, but the US president later discovers that the Russians have created a machine that will unleash an automatic nuclear strike in the event of an American attack. The rest is in the future, as they say.
Six decades later, we must rely, as humanity has in the past, on the fact that the inhabitants of the planet will retain enough sense of self-preservation not to build systems with autonomous power to destroy us all. And yet it seems reasonable to fear that Xi, Putin, Trump and Kim Jong-un themselves understand the intricacies of technology no better than I, or perhaps you.
I will never forget the story – which is historical, not just legend – of Winston Churchill, who was approached by his chief scientific adviser in August 1941. and requested permission to continue working on the process of nuclear fission with the ultimate goal of creating a bomb. This is the seed that later bore fruit through the wealth and power of the United States with the Manhattan Project.
Churchill replied that he saw no need for a more powerful weapon when the means of destruction already available seemed perfectly adequate: “I am personally quite satisfied with the explosives we have.” But because the scientists were so enthusiastic, he did not object to their research. I mean, as a Victorian born in 1874, Churchill could not have understood the epochal significance of nuclear fission, and he did not really do so until after the American bombings of Hiroshima and Nagasaki in 1945, Hastings points out. My own father wrote me a letter, days after my birth in December 1945, which he gave to me when I was 21, expecting that my generation would spend its life haunted by the specter of the Bomb. It is truly amazing how cheerfully we have spent the last 79 years, frightened by many things, but only briefly – during the Cuban Missile Crisis of 1962. – truly terrified by the prospect of intercontinental nuclear war.
Today we know this: AI will prove to be as key to world change as the creation of the atomic bomb. The book Genesis: Artificial Intelligence, Hope and the Human Spirit, co-authored by Kissinger, acknowledges, in a burst of optimism, that “If AI were to emerge as a virtually independent political, diplomatic, and military entity, it would order a change in the age-old balance of power at the expense of new, unexplored inequalities. . . Such a . . . order could witness the internal implosion of societies and the uncontrolled explosion of external conflicts.”
AI prophets predict that war will cease to be a competition between rival armies, navies, or air forces manned by humans, and instead become a struggle between rival machine systems. Courage, or the lack thereof, will cease to be a factor. The skills that matter will not be those of pilots like in the movie Top Gun, but instead of software engineers who, long before the fighting begins, program the technology that drives the battle.
The war in Ukraine has shown how missiles and unmanned aerial vehicles (drones) now dominate the battlefield. If I were in charge of arms supplies for any nation, I would refuse funding for any new combat aircraft that would not enter service until well into the 2040s that required a crew.
I remember the sense of wonderment 20 years ago in a hangar outside Kabul when I first watched a former B-52 pilot fly a drone. I sensed then, and I know now, that I had glimpsed the future.
At least the Manhattan Project was run by real people, even if some of them were as imperfect as Robert Oppenheimer. It seems right to be frightened by the prospect of machines running our future armed conflicts, not to mention managing the operations of market makers and hedge fund managers. Feldstein notes that “the prospects for accidental use of force and potential escalation are considerable.” But AI is on the horizon. We have no choice but to keep up the pace, as the Manhattan Project scientists told each other during the three years of herculean labor to create the Bomb.
We should take solace in the fact that we have survived the first eight decades of the nuclear age that many of our parents believed would be the end of us. Perhaps our children will cope with the age of artificial intelligence, which, like the Bomb, will no longer be able to simply be erased.
Comments are closed.