Nir Barzilai is a physician-scientist who built his career on a simple, awkward question: why do some people reach a hundred in good health while their contemporaries do not? His work on the genetics of exceptional longevity put centenarians at the centre of ageing research, and he has spent years arguing that ageing itself can be targeted by a drug: with metformin as the test case.
The most practically useful thing he told us is about timing. He is often quoted as saying metformin is for after fifty, and there are two reasons. The first is evidence: seven decades of research recruited people over fifty, with an average age nearer sixty-five. The second is mechanism, metformin lowers growth hormone, and in some men lowers testosterone. When you are young and still building, that is a trade you do not want.
He puts it as a rule worth carrying: “Not everything that’s good for you when you’re young is good for you when you’re old, and the other way around.” He is equally frank about the downside, three to six per cent of people get diarrhoea that does not resolve, because they lack the transporter that carries metformin into cells.
On the limits of lifespan he is careful rather than pessimistic. There is a ceiling, and the curve is starting to bend against us, but he does not believe it is permanent. And on AI he is refreshingly unimpressed by hype: he will ask ChatGPT about his own research and get a good story back, but ask it what to do next and it has nothing. “They still need me. The day they don’t need me, I’m out of here.”
He was interviewed at NUS in Singapore alongside Andrea B. Maier, and the two of them turn up in each other’s answers, including a small disagreement about whether the right verb is optimise or maximise. He concedes she is probably right.
Nir Barzilai is in our film Live Beyond: Father and Son on a Quest for Longevity and in Marek’s book Live Beyond: How to Win Your Longevity Game.

We met Nir at the National University of Singapore, where he was appearing alongside Andrea B. Maier. Two conversations came out of that visit, a joint one with Andrea, and this one. It ranges from artificial intelligence to metformin side effects, and finishes somewhere neither of us expected: on purpose, and on Aleksander’s face during the interview.
“Seventeen-, eighteen-year-olds and younger people have to know that they’re looking at their grandparents, that’s where they’re going. You get there, and you get there faster than you think.”
Nir Barzilai, Singapore
“If you ask me when it will happen, I’ll tell you, ‘By November.’ As long as you don’t ask me the year, I’ll be right.”
Nir Barzilai · Singapore, February 2024Not everything that’s good for you when you’re young is good for you when you’re old.
Nir BarzilaiI’m disappointed that it takes longer. It always takes longer than I think.
Nir BarzilaiEvery medicine that we’ll talk about has trade-offs.
Nir BarzilaiThe bigger question is always a purpose in life, right? … We have to invent it, right? And it’s good to invent it because it makes our life meaningful.
Nir BarzilaiFrom the conversation in Singapore, on artificial intelligence, on when a drug is right for you, and on what he thinks a teenager should see when they look at their grandparents.
Two reasons, and both are worth knowing. The evidence base was built on people over fifty, average age nearer sixty-five. And the mechanism cuts both ways: metformin lowers growth hormone, which is what you want when the body is breaking down and precisely what you do not want while it is still building.
“When you start breaking down, it doesn’t make sense to invest in growth.”
Singapore · from the interviewHe uses it, and he is unromantic about it. It writes a good summary of his own research; ask it what experiment to run next and it stops. His other complaint is that big data keeps rediscovering the obvious, twenty years of records on three million people to establish that obesity raises diabetes risk.
“I didn’t need AI for twenty years to know that. I’ve known it for fifty years.”
Singapore · from the interviewHis long-range picture is the most vivid image anyone gave us. In fifty years, he thinks, a seventeen- or twenty-year-old will go to the doctor and receive a treatment that erases their ageing, repeated every few months or years. They will not grow up.
“They’ll be Peter Pan.”
Singapore · from the interviewThe video interview is coming. Recorded interviews go live after the launch of the film & series Live Beyond.
From the conversation at the National University of Singapore, February 2024. Lightly edited for readability. Nir was in Singapore alongside Andrea B. Maier, who has her own page.
Nir Barzilai“When I go to ChatGPT and ask ChatGPT about my research, they’ll give me a good story. But if I say, ‘Okay, so what do I do next?’ they don’t know. They still need me. The day they don’t need me, I’m out of here.”
Nir Barzilai“For AI, you need to ask the right question. Not everybody asks the right question. There’s data on twenty years of electronic medical records of three million people. So that’s a big data set. And they discovered that obesity is a risk for diabetes. I didn’t need AI for twenty years to know that. I’ve known it for fifty years. You need to know what to ask.”
Nir Barzilai“I remember the first time that I used computational tools … and I sat with these AI guys, and I said, ‘We have to discover this thing.’ And we talked, and I talked and he talked, and we both thought that we knew what we were saying. And we said, ‘Okay, let’s meet the next week. Each one of us will write the specific aim for the grant.’ And we exchanged the grant the day before, and each one of us thought that the other was a total idiot because we didn’t understand each other.”
Nir Barzilai“Seventeen-, eighteen-year-olds look at their grandparents and say, ‘What happened to them? Where did they come from?’ I think seventeen-, eighteen-year-olds and younger people have to know that they’re looking at their grandparents, that’s where they’re going. You get there, and you get there faster than you think. And we can do something about it, and we’ll be able to do something about it, and we would look very different. By the way, I look different than my father and grandfather already. … But if they are not part of this revolution, it’s not going to work.”
Nir Barzilai“In fifty years, I think a twenty-year-old or a seventeen-year-old will come to the doctor, will get some treatment that will erase their ageing. And they’ll repeat it every few months or every few years, I don’t know. And they’ll be Peter Pan. They’ll never grow up.”
Nir Barzilai“I believe that that’s not going to be our limit. I think this limit is more to explain to people that it’s not going to be linear now. We have a roof. … We do age no matter what we do. … Life is extended, extended, and now there is a roof, and that means we’re going to start bending, and it’s going to be harder and harder to get there.”
Marek Piotrowski“When I was just talking with Brian Kennedy, also about the growth of longevity … he said that longevity scientists should talk more with the futurists because the scientists tend to be very cautious about how fast they will push the limits. The futurists are totally into the future. And then connecting these two could create a positive push.”
Nir Barzilai“I find it the opposite. I cannot tell you how many times they ask me, ‘What do you think will be in the next five and ten years?’ And I was much faster than it happened. I think that I’m disappointed that it takes longer. It always takes longer than I think. And I think it’s hard to make predictions, in particular about the future.”
Nir Barzilai“First of all, the research that has been done for seven decades always recruited people over the age of fifty. In fact, their average age was sixty or sixty-five or sometimes seventy. And so we have the data in older people. Now, as an example, one of the things that metformin is doing is lowering growth hormones, which is very important for ageing. In our centenarians, sixty per cent of them have some problems with growth hormones.”
Nir Barzilai“You need growth when you’re young. You need to be strong, you need to reproduce, you need that. … When you start breaking down, it doesn’t make sense to invest in growth. So metformin lowers growth hormone. That’s not good when you’re young. … By the way, it also, in some males, decreases testosterone. You don’t want to have low testosterone when you’re young. Not everything that’s good for you when you’re young is good for you when you’re old, and the other way around.”
Nir Barzilai“But metformin is also not tolerated. First of all, three to six per cent of people get diarrhoea that doesn’t go away. Why do they get the diarrhoea? Because they don’t have the transporter that takes metformin into the cells, and so it’s not absorbed from their intestine. … Every medicine that we’ll talk about has trade-offs, but for people who can tolerate it, it’s going to have an effect.”
Nir Barzilai“I think for you guys, exercise, nutrition, sleep, and social connectivity are going to take you a long way … Your son is interested in this conversation. He has this little smile, and he agrees with me or doesn’t agree with me.”
Nir Barzilai“I think for everyone, this is something that you can optimise. Actually, optimise is Andrea’s. I’m calling it maximise. I think she’s probably right because maximise means you can always do more, and maybe not. And maybe you shouldn’t. … The bigger question is always a purpose in life, right? … We have to invent it, right? And it’s good to invent it because it makes our life meaningful. And you have a purpose in life. Look at the trip that you’re taking. It’s just wonderful.”
Nir Barzilai sits in Biohacking Upgrades (pharma) with a real claim on Longevity Basics: asked what he wished for us, he answered exercise, nutrition, sleep and connection. Read him next to Andrea B. Maier.
Explore all 100+ expertsAlso on the Hub
Places this page mentions, each with a page of its own.