When bots have minds of their own
I’m no expert but it seems to me there might be something to this notion that artificial intelligence – otherwise known as AI – is a whole lot of trouble waiting to happen.
Even the people working to bring to market the bots of the future are sounding the alarm. Only recently, the nonprofit Center for AI Safety issued a one-sentence statement signed by 350 artificial intelligence experts, including the chief executives of the top three AI developers, warning of potential dire consequences.
The statement declares: “Mitigating the risks of extinction from AI should be a global priority alongside other societal-level risks, such as pandemics and nuclear war.”
Did you pick up on the word “extinction” in association with AI? If that doesn’t grab your attention nothing will.
One of those signing the statement was Sam Altman of powerhouse Open AI. Altman has been quoted as saying, “If this technology goes wrong, it can go quite wrong.”
Altman had called for government regulation of AI, much in the vein of the Atomic Energy Commission, which was established to keep an eye on the nuclear front. (OK, so things haven’t gone so well with that, though, as you may have noticed, we have yet to blow ourselves to bits.)
Altman recently told a Senate panel that the AI community wants “to work with the government to prevent” an AI meltdown, or more, precisely, an AI takedown of human functionality.
Months earlier, Geoffrey Hinton, the “godfather” of AI for the groundbreaking work he did on artificial intelligence at the University of Toronto as recently as 2012, raised the alarm by departing Google’s research team and then stating his belief that a future world awash with AI could be a scary place in which to live.
The 75-year-old British expat who spent the 1980s at Carnegie Mellon University told the New York Times, “It is hard to see how you can prevent bad actors from using [AI] for bad things.”
I’m supposing Hinton has in mind “bad” governments as well as “bad” private individuals and enterprises, or a combination of both. We in the United States certainly have had a taste of both, as with the fake stuff that came our way over the internet from Russia during the 2016 and 2020 presidential campaigns.
How might ordinary folks come to an understanding of AI’s potential harm? Now, I may be off base but I think maybe a hint might be found in our cars and trucks.
Do you marvel at the guidance that comes from your vehicle’s mechanism telling you the roads to take in getting from one unfamiliar place to another?
I know I do. It’s astonishing sometimes to take the routes mapped out by GPS (global positioning system). You find yourself in places you never knew existed.
I try my best to follow the spoken instructions, lest I hear the always patient admonition, “In 500 feet, make a U-turn.” It’s like hearing, “We’re going to try this again, dummy, and this time do it right!”
Therein lies a clue to a future world overawed by AI: Our willing acquiescence to technology. Our sovereign selves easily pushed aside, in effect, submerged by a disembodied digitized voice.
Of course, this is hardly the level of technology that frightens the experts. Instead, it’s AI’s capacity to learn skills by analyzing data on a scale and with a rapidity that dwarfs the capacity of the human brain.
If the trend continues, bots will know a lot more than humans. Instantly drawing on amounts of information inaccessible to even the brightest human, what’s to stop the robots of the future from brushing aside their human handlers, and in the process, plunging the world into catastrophe?
It sounds like science fiction, for sure, but these future analytical giants and consciously wonders are a source of worry.
Henry Kissinger, the former secretary of state, recently addressed the issue. “How do you build restrains into [AI] machines?” he asked. “Even today we have fighter planes that can fight … air battles without human intervention. But these are just the beginning of the process.”
David Ignatius of the Washington Post quoted Kissinger as saying, “It is the elaboration 50 years down the road that will be mind-boggling.”
Kissinger wrote a 2018 magazine piece in which he said, “Philosophically, intellectually – in every way – human society is unprepared for the rise of artificial intelligence.”
Richard Robbins lives in Uniontown. He can be reached at dick.l.robbins@gmail.com.