How Artificial Intelligence Could Transform Human Lifespan and Aging
Over the past two centuries, humans have experienced a longevity revolution. In 1824, the average life expectancy for U.S. citizens hovered around 40 years; today, that number has nearly doubled, largely due to a dramatic improvement of the infant mortality rate. Thanks to modern medicine, this increase in childhood survivability, coupled with increased life expectancy, means most of us will roam the Earth for twice as long as our great-great-great-great-grandparents did.
Of course, this amazing biological transformation has mostly occurred without one the most important tools in our modern medical research arsenal: artificial intelligence. The biggest selling point of AI is its ability to churn through unfathomable amounts of data to find novel solutions, therapies, or maybe even cures that the brightest scientific minds could never devise. Whether studying cellular senescence, telomere shortening, cancer, mitochondrial dysfunction, genomic instability, or other causes of aging and death, research has shown that AI will likely be an invaluable tool.
But what if AI was more than just a tool? What if, instead, it was the lead author of the next chapter of human longevity? This idea has led some technologists—most notably Dario Amodei, CEO of the San Francisco, California-based AI company Anthropic—to the controversial, and some would say “hilarious,” conclusion that artificial intelligence won’t just continue the longevity revolution—it’ll supercharge it.
“This might seem radical, but life expectancy increased almost 2x in the 20th century (from ~40 years to ~75), so it’s “on trend” that the “compressed 21st” would double it again to 150,” Amodei wrote in an October 2024 blog post. “There already exist drugs that increase maximum lifespan in rats by 25-50% with limited ill-effects. And some animals (e.g. some types of turtle) already live 200 years, so humans are manifestly not at some theoretical upper limit,” he continued.
By January 2025, Amodei had doubled down. At the World Economic Forum in Davos, Switzerland, he claimed that AI could double human lifespans in just five years, a simply mind-blowing timeline, but one he says is very much in line with AI development. As Amodei details in his blog post, once humans reach 150 years of age, our species could reach “escape velocity,” meaning humans could essentially choose how long they’d like to live (though he does admit this may not be “biologically possible.”)
Amodei isn’t alone with this bold prediction. Futurist Ray Kurzweil, who is best known for his predictions regarding the approaching singularity, has similarly stated that AI could push the pause button on aging as soon as 2032 in two ways. The first is through the use of AI-enabled medical nanobots that can repair damaged cells and deliver drugs directly to the affected region. The second is the ability to back up our brains to the cloud through AI—something that may not even be possible, considering we still have many questions about the human brain and how it works.
The words “revolution” and “artificial intelligence” are common bedfellows when talking about medical research. A March 2025 article in the Harvard Gazette likened the arrival of AI in medicine to the dawn of the internet or the completion of the human genome. In other words, a big freaking deal.
But using AI for screening cancer, treating disease, or developing novel therapies to improve the human healthspan—that is, increasing the amount of years one is considered healthy—doesn’t necessarily extend the species’ biological lifespan. S. Jay Olshansky, a professor of epidemiology and biostatistics at the University of Illinois Chicago, says there’s simply no evidence that AI is capable of modulating the biological process of aging. Around the same time that Amodei published his blog post, Olshansky published a study in Nature Aging effectively detailing the implausibility of a longevity revolution.
“The longevity game we’re playing today is quite different from the one we played a century ago. Now aging gets in the way, and this process is currently immutable—although scientists are working hard to find a way to modulate aging itself,” Olshansky says in an email. “The burden of proof is on the AI scientists to demonstrate they can modulate the human body and how it works. I could claim that drinking lemon water twice a day will make you live to 150, and you wouldn’t be able to prove me wrong.”
Simply put, even if AI miraculously increased the human lifespan in five years—how exactly would we know? It’ll likely take a century of verification and testing to see if these claims ever actually panned out, and AI companies have yet to show if they can extend lifespans in any measurable way. Today, humans live longer lives than at any point in our 300,000-year history, but expanding beyond that threshold will require much more than predictions and unsubstantiated promises.
“At the moment, AI is just a buzz term with no evidence that it will yield radical increases in human longevity,” Olshansky says. “Either we’re doing science or we’re not—unfortunately this field seems to draw its share of quacks.”
Darren lives in Portland, has a cat, and writes/edits about sci-fi and how our world works. You can find his previous stuff at Gizmodo and Paste if you look hard enough.
link
