Por Eve Ottenberg
 |
Photograph by Nathaniel St. Clair |
It’s time for the so-called new left to wake up to the existential threat posed by AI. The only U.S. politician cognizant of this stupendous danger is old left Bernie Sanders, who recognizes the late-capitalist AI tsunami coming for us all. He’s the only one asking questions about what happens once governments and corporations turn everything over to AI, because the only way to compete is to rely on a technology that functions at the level of “the best humans” and will soon develop beyond and surpass the best and smartest humans.
This is the overall danger posed by AI. More specific ones have been making headlines lately, namely the AI corporation Anthropic’s imbroglio with Pentagon chief Pete Hegseth. With breathtaking arrogance, Hegseth threatened to ditch Anthropic’s $200 million defense department contract for its AI called Claude, if Claude could not be used for certain despicable and frankly illegal purposes: in autonomous weapons systems that can kill people without human input and allowing Claude to be used for mass domestic surveillance. To his credit, Anthropic’s ceo Dario Amodei resisted Pentagon pressure. But these horrifying demands reveal AI’s monstrously anti-human and homicidal potential. On the evening of February 27, before Anthropic and the Pentagon could release their final positions, Trump, in a fit of pique, banned all government agencies from doing business with Anthropic. So I guess we know where the white house stands on potential abuses of AI. But in the end, this hullaballoo between an AI corporation and the Pentagon is only one aspect of the broader danger AI poses to humanity.
So what happens when, in a few years, the labor portion of the economy is zero percent and the capital portion is 100 percent? And that’s coming fast, because Donald Trump has unleashed AI. The situation might be manageable if the U.S. and China slapped some controls on this monster – so that creating AI superior to the best and smartest humans is delayed – and China would probably be willing to do that. But Trump’s U.S., particularly his venture capitalist backers? Not so much.
Most of the left regards AI as merely the latest Silicon Valley craze, like crypto. It is not. On this, the old left, represented by Sanders, is a lot smarter. Because this is not a matter of “well, I ignore AI, so it will leave me alone.” It won’t. Years ago on Twitter, The Wire’s writer, David Simon, was asked if he’d use AI to help him write, and he replied, “I’d rather put a gun in my mouth and pull the trigger.” For those of us who want AI out of our lives, that could soon be the only option.
Roughly five years ago, of the four big AI companies – Open AI, xAI, Google and Anthropic – two stated openly that if we developed an artificial general intelligence, the chance of human extinction would immediately hit 25 percent (Anthropic’s ceo DarioAmodei) or 20 percent (xAI’s boss Elon Musk). It was then still a theoretical concern, but these leaders in the field admitted that to make up for employment extermination, government would have to supply a universal basic income. But now, with the public asleep to this menace and competition having increased, those corporate bigwigs are no longer quite so honest.
Making matters worse is the fact that for Trump promoting AI keeps the stock market booming. And since Trump regards the stock market as the economy, he has issued horrible executive orders on AI, the worst being his attempts to preempt state regulation of AI, which is the only thing that might protect us from this technological Godzilla.
Three years ago, you could easily discern if you were chatting online with AI. But by winter of 2025, most AI passed the Turing Test, which means it could fool you into thinking you were chatting with a person, if it wanted to. Now if you look at more quantitative fields, like computer programming, well, in the past few months, top computer programmers have stopped writing their own code and have turned this task over to AI. In formal mathematics, AI already operates at the level of pretty good PhD students, but not yet professors. So mathematicians are still doing math. But that will likely be the first of the hard sciences that falls to total automation.
The real danger comes in the next few years, when, according to the research institute METR, the task length that AI can complete extends – because it’s been doubling every few months. So that’s an exponential trajectory that means AI will surpass most human knowledge work – computer programmers, Wall Street entry-level associates much of whose work is manipulating Excel spreadsheets (they will be merely the first to go) – in a matter of years. This change will wipe out knowledge workers, since AI will be able to perform all their tasks, faster and better.
Within Silicon Valley opinions differ on how desirable it is to summon super intelligence into the world. But Trump and his venture capitalist supporters, who have quite successfully kept this issue off the public and congressional radar, have backed those who are most gung-ho for creating super intelligence. This is dangerously short-sighted. These guys may all get rich, but they’ll do so producing a super intelligence that could easily decide in the future that we humans are a nuisance, like so many insects that need to be exterminated. If you think this is an overreaction, well then, you haven’t been paying attention.
Bernie Sanders has. Back on November 24, 2025, he released a statement entitled “The Future of AI and Its Impact on Humanity.” “The question is: Who will control this technology? Who will benefit from it? And who will be left behind?” Then on February 20, the Guardian reported that Sanders “Warned that Congress and the American public have ‘not a clue’ about the scale and speed of the coming AI revolution, pressing for urgent policy action to ‘slow this thing down.’” Sanders calls this the “most dangerous moment in the modern history of this country.” He added, “the Congress and the American people are very unprepared for the tsunami that is coming.” Sanders reissued his call for a moratorium on the expansion of AI data centers to “slow down the revolution and protect workers.”
Sanders first called for a moratorium in December 2025. “If there are no jobs and humans won’t be needed for most things, how do people get an income to feed their families, to get healthcare or to pay rent? There has not been one serious word of discussion in the Congress about that reality.” As the Guardian reported December 28, 2025, he questioned the motives of “the richest people in the world” pushing AI. “He singled out tech moguls Elon Musk, Mark Zuckerberg, Jeff Bezos and Peter Thiel.” Sanders calls for slowing “this process down. It’s not good enough for the oligarchs to tell us, it’s coming, you adapt…They’re going to guarantee health care to all people? What are they going to do when people have no jobs…make housing free?” Again and again, Sanders says, correctly, we need to slow this hurricane down.
So when it comes to AI, old left Bernie Sanders knows the score. Donald Trump does not. Or, more likely, he just doesn’t care. In 2025, Trump issued four executive orders promoting AI and deregulating it. Worst of all was his December 2025 order designed to stop states from restricting AI. Scheduled for March 2026, the Secretary of Commerce is tasked with identifying burdens on AI, among other things. Trump’s assault on the only guard-rail against explosive AI development, namely state regulations literally threatens us all. The only way out of this dead end that AI is corralling us into lies in Congress. If Sanders can wake up his fellow congress members, then we humans might stand a chance. Otherwise, it’s a future of no jobs, no income and the likelihood that super-intelligent AI comes to regard the human species as redundant or outright expendable.
Eve Ottenberg is a novelist and journalist. Her latest novel is Booby Prize. She can be reached at her website.
Nenhum comentário:
Postar um comentário