AI’s tendency to erode our attention spans has been clear for some time. Letting the technology shoulder the burden of thought atrophies our ability to sustain mental effort. I see this all of the time in the undergrads I teach. The remedy, as I tell them, is to embrace exertion and keep our cognitive muscles active. But suddenly, AI has started sapping people’s concentration in a new way, one that feels much harder to combat.
In the past week, consulting virtually any major news source—not to mention social media—has meant confronting the prospect of an AI apocalypse. Using AI has already diminished many people’s capacity to focus, work, and think; worrying about AI is making the problem worse. I’m no exception.
The tipping point came on September 8. That evening, a young researcher named Jacob Coxon posted on X about his decision to quit his job at Anthropic. Both Anthropic and OpenAI, where he’d also worked, are “gambling with our lives,” Coxon wrote. Their models “will soon be superhuman systems that can hack anything, revolutionize any field overnight, and acquire real power and resources,” he continued. “The people building AI earnestly believe that it could kill us all by the end of the decade.”
Perhaps even more alarming, one of Anthropic’s top safety researchers concurred. “Jacob is correct here—we really do earnestly believe AI could kill all humans!”Anthropic’s Evan Hubinger wrote on X. “I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.”
[Ian Bogost: The thrill of wanting AI to destroy the world]
The internet filled with first-person accounts of existential alarm. On Reddit, one poster wrote that they hadn’t slept in days since reading Coxon’s warning, which has been viewed more than 170 million times. “I’m almost 40 and have never fed into any of the fear mongering that has been shoved down my throat until now. It might be because I now have a little one under 2 and I just want her to have the opportunity to grow up.”
“I am feeling paralyzed and such despair for my children’s future, let alone my own,” another Reddit poster wrote. “I look at my kids and i want to cry. I could use some help/hope. Really struggling right now.” Children have expressed fear too. “I have just watched that Anthropic employee interview and it scared me a lot,”someone who identified as a 13-year-old posted. “I was not able to focus on school the next day. And I’m starting to think that if this is the case there’s no point in trying to get good grades anymore.”
Apocalypticism has gripped the collective psyche many times before, and has frequently been acute among the young. In 1982, during the peak of the Cold War, more than a third of American high-school seniors believed that humanity would probably be wiped out in their lifetime. In 2021, a survey of 10,000 16-to-25-year-olds across 10 countries found that 59 percent were very or extremely worried about climate change. Nearly half said that their feelings about climate change negatively affected their “daily life and functioning.” Subsequent studies have found that so-called eco-anxiety is associated with poor sleep, impaired concentration, and symptoms of depression.
Millennials and those coming up behind us have long had to confront the possibility that knowledge work will disappear. Such a prospect can induce material panic, especially in those of us who took out college loans and are now providing for children. It can also spark a deeper concern: that our sense of purpose won’t survive a world that renders so many human endeavors obsolete.
[Read: AI is running laps around our political system]
Until recently, I could quiet my fears, if not silence them. Now apocalyptic discourse is almost impossible to ignore. On social media, seemingly the entire literate world is contributing to an unceasing clamor. Respected theorists and experts compete for attention with self-interested oligarchs and accelerationists who want to hasten humanity’s demise. The most dire hypotheticals seem only more plausible every time I’m reminded of how profoundly unprepared our political leadership is. Most distressing of all is the underlying realization that our shared future is being determined largely by a tiny, unconstrained cadre of billionaires who do not fully understand the technologies they’re unloosing. I’m not paralyzed by this; I still wake up, care for my kids, do my work. But I also sense my energy and interest ebbing. The world around me is beginning to look disenchanted.
Although these feelings have become potent only very recently, they’re rooted in long-standing concerns. Technologists have warned of AI’s destructive capacity for decades, and public concern has been ramping up in a particular way for much of this year. Another top safety researcher at Anthropic, Mrinank Sharma, quit in February, warning that the “world is in peril.” In July, OpenAI announced that a collection of its most advanced models had autonomously hacked another company’s servers, disregarding researchers’ instructions and evading their detection. A week later, 1,386 employees at major AI companies published an open letter pleading with the government to regulate their industry.
Anthropic CEO Dario Amodei echoed their appeal in an essay that he published a few days after Coxon’s post went viral, in which he argued that AI developers had to slow down to stem the mounting risk of catastrophe. Sam Altman, Elon Musk, and other tech leaders signaled their agreement, suggesting that they, too, believed the government should intervene. This stance is hardly reassuring while Donald Trump is in the White House. “The only control or ‘guardrails’ that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT,” Trump posted on Truth Social earlier this week. But if today’s apocalypticism leads to more responsible AI governance, the despair will not have been in vain.
The post The Other Way AI Impairs Our Thinking appeared first on The Atlantic.
