Playback speed
Share post
Share post at current time

Tony Blair - Life of a PM, The Deep State, Lee Kuan Yew, & AI's 1914 Moment

What makes a good leader?

I chatted with former Prime Minister Tony Blair about:

  • What he tells the dozens of world leaders who come seek advice from him

  • What he learned from Lee Kuan Yew

  • Intelligence agencies track record on Iraq & Ukraine

  • How much of a PM’s time is actually spent governing

  • What will AI’s July 1914 moment look like from inside the cabinet


Watch on YouTube. Listen on Apple PodcastsSpotify, or any other podcast platform. Read the full transcript here.

Follow me on Twitter for updates on future episodes.


  • Prelude Security is the world’s leading cyber threat management automation platform. Prelude Detect quickly transforms threat intelligence into validated protections so organizations can know with certainty that their defenses will protect them against the latest threats. Prelude is backed by Sequoia Capital, Insight Partners, The MITRE Corporation, CrowdStrike, and other leading investors. Learn more here.

  • This episode is brought to you by Stripe, financial infrastructure for the internet. Millions of companies from Anthropic to Amazon use Stripe to accept payments, automate financial processes and grow their revenue.

If you’re interested in advertising on the podcast, check out this page.


(00:00:00) – A prime minister’s constraints

(00:04:12) – CEOs vs. politicians

(00:10:31) – COVID, AI, & how government deals with crisis

(00:21:24) – Learning from Lee Kuan Yew

(00:27:37) – Foreign policy & intelligence

(00:31:12) – How much leadership actually matters

(00:35:34) – Private vs. public tech

(00:39:14) – Advising global leaders

(00:46:45) – The unipolar moment in the 90s


00:00:00 – A prime minister’s constraints

Dwarkesh Patel 00:00:00

Today I have the pleasure of speaking with Tony Blair who was Prime Minister of the UK from 1997 to 2007 and now leads the Tony Blair Institute, which advises dozens of governments on improving governance, reform, and adding technology.

For my first question, I want to go back to your time in office when you first got in. You had these large majorities. What are the constraints on a prime minister, despite the fact that they have these large majorities? Was it the other members of your party fighting against you? Was it the deep state? What part was constraining you at that point?

Tony Blair 00:00:35

The biggest constraint is that politics, in particular political leadership, is probably the only walk of life in which someone is put into an immensely powerful and important position with absolutely zero qualifications or experience. I'd never had a ministerial appointment before. My one and only one was being prime minister. It’s great if you want to start at the top, but it's that that’s most difficult.

When you're running for office, you have to be the great persuader. The moment you get into office, you really have to be the great chief executive. Those two skill sets are completely different. A lot of political leaders fail because they've failed to make the transition.

Those executive skills — focus, prioritization, good policy, building the right team of people who can actually help you govern — are crucial. The moment you become the government, the saying becomes less important than the doing. Whereas when you're in opposition or running for office, it's all about saying.

All of these things mean that it's a much more difficult, much more focused role. Suddenly you're thrust into this completely new environment when you come in. That's what makes it the hardest thing. Of course, you do have a situation with the system as a system. It's not that there's this great deep state theory. We can talk about that but that's not the problem with government.

The problem with government is not that it's a conspiracy, either left-wing or right-wing. It's a conspiracy for inertia. The thing about government systems is that they always think, "we're permanent, you've come in as the elected politician, you're temporary. We know how to do this and if you only just let us alone, we would carry on managing the status quo in the right way." That's the toughest thing, making that transition.

Dwarkesh Patel 00:02:48

That's really interesting. Let’s take you back with everything you knew, let's say in 2007, but you have the majorities and the popularity you had in 1997. Is it that you know now what is a waste of time? Would you say, "I'm not going to do these PMQs, they're total theatrics" or "I'm not going to meet the queen"? Is it the time? Is it that you're going to go against the bureaucracy and say, "I think you're wrong about your inertia"? What fundamentally changes?

Tony Blair 00:03:20

It wouldn't be that you wouldn't do Prime Minister's Questions, because Parliament will insist on that. You certainly wouldn't want to offend the Queen, who was the monarch in my time. But you're right. You would have a much clearer idea of how to give direction to the bureaucracy and how to bring in outside skilled people who can help you deliver change.

I always split my premiership into the first five years, which in some ways were the easiest. We were doing things that were important, like a minimum wage. We did big devolution. We did the Good Friday peace agreement in Northern Ireland.

It was only really in the second half of my premiership that we started to reform healthcare, education, and criminal justice. That's when your skill set as a chief executive really comes into play.

00:04:12 – CEOs vs. politicians

Dwarkesh Patel 00:04:12

Many people have a preconception that if you could get a successful CEO or businessperson into office, these executive skills would transfer over pretty well into becoming a head of state. Is that true? If not, what is it that they'd be lacking that you need to be an electable leader?

Tony Blair 00:04:28

This is really interesting and I think a lot about this. The truth is those skills would transfer to being a political leader. They're not the only skills you need because you still have to be a political leader. Therefore, you've got to know how to manage your party. You've got to know how to frame certain things.

As a CEO of a company, you're the person in charge. You can more or less lay down the law. Politics is more complicated than that. When highly skilled CEOs come into politics, oftentimes they don't succeed. That's not because their executive skill set is the problem. It's because they haven't developed a political skill set.

Dwarkesh Patel 00:05:11

I was reading your memoir. There were a couple of times where you said that you realized later on that you had more leverage.  In retrospect, you were able to do things that you didn't do at the time.

Is that one of the things that would change if you went back to 1997? Would you realize, "I actually can fire this entire team if I don't think they're doing a good job. I actually can cancel my meetings with ambassadors." Did you have more leverage than you realized at the time?

Tony Blair 00:05:39

In terms of running the system, yes. Definitely with the benefit of experience, I would have given much clearer directions. I would have moved people much faster. Again, in politics this is where it's different from running a company. In a company, by and large, you can put the people in the places you want them. Except in exceptional circumstances, this is true. If you're running a company, you've got no one in the senior management who you don't want to be in the senior management.

Politics isn't like that because you've got political elements you may have to pacify. There may be people that you don’t particularly want because they’re not particularly good at being ministers. However, they may be very good at managing your party or your government in order to get things through.

What I learned over time is that the important thing is to put quality people into the core positions that really matter to you. Don't fall short on quality. It's one of the really interesting things. Being a political leader is the same as leading a company, or a community center, or a football team. It all comes to the same thing. It's such an obvious thing to say that it's all about the people. But it's all about the people.

You get really good, strong, determined people who share your vision and are prepared to get behind and really push. You actually don't need that many of them to change your country. But they do need to be there.

Dwarkesh Patel 00:07:23

That's really interesting. This isn’t particular to the UK but even in Western governments, people are often frustrated by this. They elect somebody they think is a change maker, but things don't necessarily change that much. The system feels an inertia. If this is the case, is it because they didn't have the right team around them?

If you think of Obama or Trump or Biden, at the very top I assume they can recruit top people. To the extent that they weren't able to exact the change they wanted, it wasn’t because they didn't get the right chief of staff, right? They can probably get the right chief of staff.

Tony Blair 00:07:58

Absolutely. They can get really good people. One of the things you actually learn about being at the top of a government is that if you pick up the phone and say, "I need you to come and help," to someone, pretty much they will come.

That’s not the problem. I say this often to the leaders that I work with. We work in roughly 40 different countries in the world today and that's only growing. We have teams of people that go and live and work alongside the president's team. I talk and exchange views with the president or the prime minister. Very often, the two problems are these.

Number one, people confuse ambitions with policies. Often I will speak to a leader and say, "so what are your policies?" They'll give me a list of things. I say to them, "those aren't really policies, they're just ambitions." Ambitions in politics are very easy to have, because they're just general expressions of good intention.

The problem comes with the second challenge. Though politics at one level is very crude — you're shaking hands, kissing babies, making speeches, devising slogans, attacking your opponents — when it comes to policy it's a really intellectual business. If they've only got ambitions, then they haven't really undertaken the intellectual exercise to turn those into policies. Policies are hard. It's hard to work out what the right policy is.

Take this AI revolution. We're living through a period of massive change. This is for sure the biggest technological change since the Industrial Revolution. This is really difficult work, for political leaders today to understand that, to work out what the right policy is, to access the opportunities, mitigate the risks, and regulate it.

What happens a lot of the time is that people are elected on the basis that they are change makers because they've articulated a general vision for change. But when you then come to, "okay, what does that really mean in specific terms?" That's where the hard work hasn't been done. If you don't do that hard work and really dig deep, then what you end up with are just ambitions. They remain ambitions.

00:10:31 – COVID, AI, & How Gov. Deals with Crisis

Dwarkesh Patel 00:10:31

Now that you've brought up AI, I want to ask about this. I do a lot of episodes on AI. To the people who are in the industry it seems plausible, though potentially unlikely, that in the next few years you could have a huge July 1914-type moment, but for AI. There's a big crisis. Something major has happened in terms of misuse or a warning shot.

How well would today’s governments deal with this, given how they function either in the West or how you see them function with the other leaders you advise? They get this news about some AI that's escaped or some bioweapon that's been made because of AI. Would it immediately kick off a race dynamic from the West to China? Would they have the technical competence to deal with this? How would that shape out?

Tony Blair 00:11:17

Right now, definitely not. One of the things that we do as an institute, and one of the reasons I'm here in Silicon Valley, is to try and bridge the gap between what I call the change makers and the policy makers. A lot of the time the policy makers just fear the change makers. The change makers don't really want anything to do with the policy makers because they just think they get in the way. You don't have a dialogue.

Let’s say what you're describing were to happen. By the way it's possible at some point it does happen. If it happened right now, political leaders wouldn't know where to begin in solving that problem, or what it might mean.

This is what I keep saying to the political leaders I'm talking to today. We're likely to have a change of government in the UK this year. I am constantly saying to my own party, the Labour Party, which will probably win this election, "you've got to focus on this technology revolution." It's not an afterthought. It's the single biggest thing that's happening in the world today. It’s of a real world nature and is going to change everything.

Leave aside all the geopolitics and the conflicts and war and America, China, all the rest of it. This revolution is going to change everything about our society, our economy, the way we live, the way we interact with each other. If you don't get across it, then when there is a crisis like the one you're positing, you're going to find that you have no idea how to deal with it.

Dwarkesh Patel 00:12:54

COVID is maybe a good case study to analyze how these systems function. The Tony Blair Institute made what were very sensible recommendations to governments, many of which went unheeded. What I thought was especially alarming about COVID was not only that governments made mistakes with vaccine rollout and testing and so forth, but that these mistakes were so correlated across major governments. No Western government, maybe no government, got COVID right.

What is the fundamental source of that correlation? In the way that governments are bad at dealing with crises, they seem in some correlated way bad with crises. Is it because the same people are running these governments? Is it because the apparatus is the same? Why is that?

Tony Blair 00:13:40

First of all, to be fair to people who were in government at the time of COVID, it was a difficult thing to deal with. I always said the problem with COVID was that it was plainly more serious than your average flu, but it wasn't the bubonic plague.

To begin with, there was one very difficult question. To what degree do you try and shut the place down in order to get rid of the disease? You had various approaches to that, but that's one very difficult question. Most governments try to strike a middle course, to do restrictions but then ease them up over time.

Then you have the issue of vaccination. Normally with drugs it takes you years to trial a drug. You had to accelerate all of that. That was done to be fair. But then you have to distribute it. That is also a major challenge.

Part of the problem was that governments weren't sure where to go for advice. They had scientific advice and medical advice. They had to balance that with the needs of their economy and the anxiety a lot of people had. They felt that when you were having a large shutdown, they were going to be hugely disadvantaged as indeed people were. There's an argument for saying for the developing world that lockdowns probably did more harm than good.

Dwarkesh Patel 00:15:15

It sounds like you're saying that they made the trade-off you're describing in the wrong way. They could have gone heavier on the testing and vaccination rollout so that the lockdowns could have been avoided and fundamentally more people's lives could have been saved.

In some fundamental sense, the pandemic is a simpler problem to deal with than an AI crisis in a technical sense. Yes, you have to fast-track the vaccines, but it's a thing we've dealt with before. There are vaccines and you roll them out. If the government can't get that right, how worried should we be about their ability to deal with AI risk?

Should we then just be fundamentally averse to a government-led answer to the solution? Should we hope the private sector can solve this because the government was so bad at COVID?

Tony Blair 00:15:57

What the private sector can do is to input into the public sector. In COVID, the countries that handed vaccine production to the private sector and said, "run with it" are the countries that did best, frankly.

Especially with something as technically complex as AI, you are going to rely on the private sector for the facts of what is happening and to establish the options about what you do. But in the end, the government or the public sector will have to decide which option to take.

The thing that makes governing difficult is that when you decide, you divide. The moment you take a decision on a public policy question, there are always two ways you can go. With COVID, you could have decided to do what Sweden did and let the disease run pretty much. You could have decided to do what China did and lock down completely. What happened with China was that once we got the Omicron variant, it became obvious you weren't going to be able to keep COVID out. They didn't have the facility or the agility to go and change policy.

These policy questions are hard. It's very easy with hindsight to say, "yeah, you should have done this, should have done that." If this happened in relation to AI, you would absolutely depend on the people who were developing AI to be able to know what decision you should take, not necessarily how you decide it, but what the choice is.

Dwarkesh Patel 00:17:44

Before COVID we had these health bureaucracies and we hoped they functioned well. It turns out many of them didn't. We probably have equivalents in terms of AI. We have government departments that deal with technology and commerce and so forth. What if they're as potentially broken as we found out that much of our health bureaucracy was?

If you were prime minister now or for the next government, what would you do other than "we're going to have a task force and we're going to make sure we're making good decisions"? I'm sure people were trying to make sure the CDC was functional and it wasn't. Would you just fire everybody there and make a new department? How would you go about making sure it's really ready for the AI crisis?

Tony Blair 00:18:31

You've got to distinguish between two separate things. One is making the system have the skills and the sensitivity to know the different contours of the crisis that you've got. The other is being able to produce potential solutions for what you do.

We needed to rely upon the scientific community to say, "this is how we think the disease is going to run." We relied on the private sectors to say, "this is how we could develop vaccines." We had to rely on different agencies in order to say, "I think you could concertina the trial period to get the drugs."

In the end, the decision of whether you lock down or you don't lock down, you can't really leave it to those people because that's not their expertise. If you look at it at the moment, some people want to regulate AI now on the basis that it's going to cause enormous dangers and problems.

Because it's general purpose technology, there are real risks and problems associated with it. Europe is already moving in quite an adverse regulatory way. On the other hand, there will be people who say, "if you do that, you're going to stifle innovation, and we're going to lose the opportunities that come with this new technology."

Balancing those two things, that's what politics is about. You need the experts, the people who know what they're talking about, to tell you "this is how AI is going to be. This is what it can do. This is what it can't do. We are explaining the technicality to you." But ultimately, what your policy is, you've got to decide that. By the way, whichever way you decide it, someone's going to attack you for it.

00:21:24 – Learning from Lee Kuan Yew

Dwarkesh Patel 00:21:24

Let's go back to the topics discussed with foreign leaders at TBI. Take Lee Kuan Yew and his position in the 1960s and the Singapore he inherited. If you were advising Lee Kuan Yew with the advice you would likely give to a developing country now, would Singapore have been even more successful than it ended up? Would it have been less successful? What would the effect of your advice now have been on Singapore in the 1960s?

Tony Blair 00:21:51

With Lee Kuan Yew in Singapore it's the wrong way around. I learned so much from him. I first went to see him in Singapore back in the 1990s when I was leader of the Labour Party. The Labour Party had been really critical of Singapore. The first thing he said to me when I came into the room was, "why are you seeing me? Your party's always hated me." I said, "I want to see you because I've watched what you do in government and I want to learn from it."

I don't think there's anything I could have told Lee Kuan Yew. He's a fascinating leader. This is the interesting thing about government. Don’t look upon government as a branch of politics. Look upon it as its own discipline, a professional discipline. You can learn lessons of what's worked and what doesn't work.

The fascinating thing about Lee Kuan Yew is that right at the beginning he took three important decisions for Singapore. Each one of them now seems obvious. Each one at the time was deeply contested.

Number one, he said, "everyone's going to speak English in Singapore." There were lots of people who said to him at the time, "no, we've been thrown out of Malaysia effectively. We're now a fledgling country, a city-state country. We need to have our own local language. We need to be true to our roots and everything." He said, "no, English is the language of the world and we're all going to speak English." Today that's what happens in Singapore.

Secondly, he said, "we're going to get the best intellectual capital and management capital, from wherever it exists in the world, and we're going to bring it to Singapore." Again, people said, "no, we should stand on our own two feet" and "you're also bringing in the British who we've got all these disputes with." He said, "no, I'm going to bring in the best from wherever they are, and they're going to come to Singapore." Today, Singapore exports intellectual capital.

The third thing he did was he said, "there's going to be no corruption. And one of the ways we're going to do that is we're going to make sure political leaders are well paid.” The Singaporean leaders are the best paid in the world by a factor of about 10 for the next person. And he said, “there's going to be zero tolerance of corruption, just zero tolerance of corruption.” Those are the three decisions that were instrumental in building Singapore today.

Dwarkesh Patel 00:24:19

If you could go to the UK or the US right now and narrow down the list to three key priorities, what could somebody in power do to fix them? Let’s say Starmer is elected in the UK. We’ll see whoever becomes president in the US.

Do Western leaders have the power to enact what the equivalent would be for their societies right now? Or do you have all this inertia and you can't start fresh like Singapore could in the 1960s?

Tony Blair 00:24:53

No, you definitely can. The American system's different because it's a federal system. In many ways, it's a good thing that there are limits to what the federal government can do in the U.S.

If you take the UK or most governments, there's a lot of power at the center. We've just been talking about the technology revolution. How do you use it to transform healthcare, education, and the way government functions? How do you help educate the private sector as to how they can embrace AI in order to improve productivity?

This is a huge agenda for a government and a really exciting one. I keep saying this to people that are in politics today. Sometimes people get a bit depressed about being in politics because you have all this criticism. Certainly people in the West feel that society's not changing fast enough and well enough. I say no, it's a really exciting time to be in politics because you've got this massive revolution that you've got to come to terms with.

Dwarkesh Patel 00:25:53

Speaking of federalism, you advise dozens of governments. For any one leader you're probably giving very sensible advice for their country. It's a positive expected value. To the extent that limits variance and experimentation across countries in different ways to govern or different policies, are we losing the ability to discover a new Singapore? Western NGOs and other global institutions will give them good recommendations but is there maybe something missing that we don't understand that an experimentation would reveal?

Tony Blair 00:26:31

We really don't do that with our governments. One of the things governments should be able to do is experiment to a degree. Part of the problem with systems is there's always a bias towards caution. That's what I mean by saying that with systems, if there were conspiracy for anything, it's for inertia.

What we do is we concentrate with governments on what is true no matter what government you're in. I describe four Ps of government when you get into power.

Number one, you've got to prioritize because if you try to do everything, you'll do nothing.

Number two, you've got to get the right policy, what we were talking about before. That means going deep and getting the right answer. That often means bringing people in from the outside, who can tell you what the right answer is. That has nothing to do with left or right, it's usually to do with practicality.

Number three, you've got to have the right personnel.

Number four, you've got to performance manage. Once you've decided something and you've got a policy, you've got to focus on the implementation.

Whether you're running the United States of America or you're running a small African country, those things are always true.

00:27:37 – Foreign Policy & Intelligence

Dwarkesh Patel 00:27:37

Let's talk about foreign policy for a second. This is not just you but every administration, especially Western administrations, has to deal with these irascible dictatorial regimes. They're right on the brink of WMDs. They make all these demands in order to put off their path towards WMDs.

Obviously you had to deal with Saddam. Today we have to deal with Iran and North Korea. It seems like sanctions don't seem to work. Regime change is really expensive. Is there any fundamental solution to this kind of dilemma that we keep being put into decade after decade? Can we just buy them a nice mansion in Costa Rica? What can we do about these kinds of regimes?

Tony Blair 00:28:20

It's very difficult. Take Iran today. There's no appetite in the West to go and enforce regime change. You could do two things that are really important. Iran is basically the origin of most of the destabilization across the Middle East region and beyond. First of all, you can constrain it as much as possible. Secondly, you can build alliances. That means that their ability to impact is reduced.

It's a constant problem because they're determined to acquire nuclear weapons capability. We want to stop them doing that. We don't want to engage in regime change. On the other hand, all the other things that you do will be limited in their effect. So it's difficult. It's very difficult, particularly now that you have an alliance that has grown up. China, Russia, Iran, and, to a degree North Korea, work closely together.

Dwarkesh Patel 00:29:34

As a leader, how do you distinguish among cases when the intelligence communities come to you? How do you distinguish between a case like Iraq, where potentially they got it wrong, versus Ukraine, where it seems like they were on the ball? How do you know which intelligence to trust? And how good is Western intelligence generally? How good are the Five Eyes?

Tony Blair 00:29:53

Generally, it's extremely good. The Five Eyes is extremely good.

Dwarkesh Patel 00:29:57

And how do you distinguish the cases where they're not?

Tony Blair 00:30:00

It's difficult. With the benefit of hindsight, particularly in relation to Iraq, you've got to go much deeper. You've got to not take the fact that there were all these problems in the past as an indication of what's happening now or in the future. On the whole, Western intelligence is reasonably good. Of course, we'll get much better now with the tools it's got at its disposal.

Dwarkesh Patel 00:30:29

How much situational awareness do they have about the topics you're talking about, whether it's AI or the next pandemic? I mean these forward-looking kinds of problems rather than who's doing an invasion? Maybe they have a lot of expertise and decades of experience with the latter sort. Predicting who's got the data center where and those kinds of things, how good are they at that?

Tony Blair 00:30:47

They're all over this stuff now, the intelligence services here in America and in the UK. You’ve also got a whole new category of threat to deal with because the cyber threats are real and potentially devastating in their impact. You can see from Ukraine, war is going to be fought in a completely different way in the future as well.

00:31:12 – How much leadership actually matters

Dwarkesh Patel 00:31:12

I want to ask you about your experience advising leaders, or interacting with them in office, and seeing how their countries progressed. How much of the variance in outcomes for countries is explained by the quality of the leadership versus other endogenous factors such as human capital, geography, etc.?

Tony Blair 00:31:30

The whole reason I started this institute was because the quality of governance — of which leadership is a big part — is the determinant. In today's world where capital's mobile, technology's mobile, any country with good leadership can make a success. If you take two countries side by side — same resources, same opportunities, same potential therefore — one succeeds and  one fails. If you look at it, it's always about the quality of decision making.

Before the Ukraine War, when both Poland and Ukraine came out of the Soviet Union in the early 1990s, people would have given Ukraine as much chance as Poland of doing well. Poland today is doing really well. Why is that? Because they've joined the European Union. They've had to make huge changes and reforms. Therefore they're a successful country.

Look at Rwanda and Burundi. It was Rwanda that suffered the genocide, but Rwanda today is one of the most respected countries in Africa. Look at the Korean peninsula. It’s the biggest experiment in human governance that’s ever been: North Korea and South Korea. South Korea had the same GDP per head as Sierra Leone in the 1960s. Now it's one of the top countries in the world.

Dwarkesh Patel 00:32:54 

Do you think that's fundamentally a question of who the leadership was, like Park and South Korea? There are other factors that are obviously different between these countries. You also mentioned that leadership determines governance. But if you look at a system like the US or the UK, we've had good and bad leaders. Fundamentally, the quality of the governance doesn't seem to shift that much between who the leader is. Is the quality of governance and the quality of the institutions, separate endogenous variables from the leadership?

Tony Blair 00:33:28

The institutions matter and good leaders should be able to build good institutions. We were talking about Lee Kuan Yew in Singapore. Would Singapore be Singapore without those decisions that he took? No.

Take, for example, China and Deng Xiaoping. When he decided to switch after the death of Mao and switch policy completely to open China up, that made the difference.

If you track back India's development over the past 25 years, you can see the points at which decisions were taken that gave India the chance it has today. The interesting thing about this is how much it really does matter who the leader is and the governance of the country.

What I'd say to people engaged in this groundbreaking revolution of artificial intelligence is: we need your help in changing government and changing countries. When I'm talking to people and leaders we work with in the developing world today, I say to them, "don't try and replicate the systems of the West." You don't need to do that. You can teach your children better and differently without building the same type of system we have in the West. I wouldn't design the healthcare system in the UK today as it is now, if we had the benefit of generative AI.

The leaders that are going to succeed in these next years will be the people that can understand what is happening in places like this. Here’s the frustrating thing from our perspective, from the perspective of leaders. Even though they would probably be very well intentioned towards the developing world, people in the technology sector think, "I don't know what I can do in order to help.” Actually there's massive amounts they can do in order to help.

00:35:34 – Private vs. public tech

Dwarkesh Patel 00:35:34

When you look at the IT revolution and how much it's improved, say market services versus how much it's improved public sector services, there's clearly been a big difference. Would it have just been better to privatize the things that IT could have enabled more of, like education? What lessons does that have for AI? The public sector didn't seem that good at integrating IT. Maybe it’ll be bad at integrating AI. Should we, say, just privatize healthcare and education as much as possible because all the productivity gains will come from the private sector of those things anyways?

Tony Blair 00:36:15

It's a great question. It's the single most difficult thing. You can't just hand everything over to the private sector because in the end, the public will expect the government to take account of the public interest. You may say that the government's useless at protecting the public interest. That's another matter. On the whole, people in America and the UK aren't going to say, "okay, just hand it over to these tech giants and let them run everything."

We have a whole program in my institute now, which we call the reimagined state. There was a minimalist state in the 18th century, and in the first part of the 19th century. That grew in the last part of the 19th century, and first part of the 20th century, into a maximalist state. You look for government to do a lot of things for you and the state grows large.

We should reimagine the state today as a result of this technology revolution and make it much more strategic. It's much more about setting a framework and then allowing much more diversity, competition. The hardest thing about the public sector in those circumstances is creating self-perpetuating innovation. If you don't innovate in the private sector, you go out of business. If you don't innovate in the public sector, you're still there. It's just the service has got worse. This is the really tough intellectual task. 

For example, in education today in America you'll have a significant tail of kids that are taught really badly. It’s probably the same in any Western country. No one today should be taught badly. By the way, everyone should be taught on an individual basis, on a personalized basis. For example, look at what Sal Khan's doing with Khan Academy. We work with them. There are other people doing great things in education using technology.

We should be able to create a situation in which young people today are able to learn at the pace that is good for them. No young person should be without opportunity. How you reform the system to allow that to happen, that's the big challenge. With the healthcare system, you will in time end up with an AI doctor. You'll end up with an AI tutor.

The question will be what's the framework within which those things operate. How do we use them to allow better service? How do we allow a lot of the people within the healthcare or education systems to concentrate on the most important part of their learning? For example, if you're a doctor now, you’re having to write up a whole lot of notes after a consultation or do lesson planning if you're a teacher.

00:39:14 – Advising leaders and implementing policy change

Dwarkesh Patel 00:39:14

Let’s go back to TBI for a second. When you give a leader some sensible advice and then they don't follow through on it, what usually is the reason that happens? Is it because it's not politically palatable in their country? Is it because they don't get it? To the extent that you have good advice that’s ignored, why does that happen?

Tony Blair 00:39:34

It happens usually for two reasons. Number one, it's really hard to make change. What I learned about making change is that there's a certain rhythm to it. When you first propose a reform, people tell you it's a terrible idea. When you're doing it, it's absolute hell. After you've done it, you wish you'd done more of it.

Sometimes people just find the system too resistant. There might be vested interests that get in the way of it. I sometimes come across countries that are island states. They have warm weather, but they get all their electricity from heavy fuel oil when they've got limitless amounts of solar and wind that they could be using. There’s a vested interest.

The other thing about the government is that it’s a conspiracy of distraction. You've got events and crises and scandals. The most difficult thing is to keep focused when you've got so many things that are diverting you from that core task. Sometimes what we do with our leaders is we say to them, "okay, we're going to do an analysis. Here are your priorities. Here's how much time you spend on them." You end up literally with people spending 4 percent of their time on their priorities. You say, "no wonder you're not succeeding."

Dwarkesh Patel 00:42:08

I'm not necessarily picking on your time, but for any head of state in Western government or any government. How much of the time they spend is fundamentally wasted in the sense of the three key priorities you would have, say, identified for Singapore in the 1960s? Time that’s spent that’s not fundamentally moving the ball forward on things like that. It’s like meeting people, ambassadors, press, whatever. How much of the time is just that?

Tony Blair 00:42:40

A lot.

Dwarkesh Patel 00:42:41

Greater than 80%? 90%?

Tony Blair 00:42:43

No, that would be too high. It’s a lot and a lot more today.

I often say this to leaders and I think this is the single biggest problem with Western politics today. When my kids were younger, I would say to them, "work hard, play hard.” Work hard, play hard equals possibility of success. Play hard, work hard is a certain failure. You'll play so hard you never end up working hard.

The equivalent in politics is policy first, politics second. In other words, work out what the right answer is, and then work out how you shape the politics around that. But what actually happens in a lot of systems today is politics first and policy second. People end up with a position that they’ve chosen for political reasons, and then they try to shape a policy around that politics. It never works.

The most important thing, for policymakers and intellectual business, is that you've got to get the right answer. There is a right answer, by the way. Often, the reason why it's so difficult to govern today is there's so much political noise. It's hard to get out of that zone of noise and sit in a room with some people who really know what they're talking about and go into the detail of what is the solution to a problem.

Sometimes when I talk to leaders about this, I find that they just say to me, "I don't have the time to do that." I say, "if you don't have the time to do that, you are going to fail. Because in the end, you won't have the right answer.” You've got to believe that over time, the best policy is the best politics.

Dwarkesh Patel 00:44:47

It's not just the time you had to spend, for example talking to the press, but also the kinds of topics that draw attention. There’s some statement your cabinet minister made on the BBC and some latest scandal.

As you wrote in your book, the 30 minutes of the PMQs is not the big deal. It's the two days you spent in anxiety and preparation. Fundamentally, is the attention distraction the bigger issue than the actual time you spend on these events?

Tony Blair 00:45:17

It is to a degree. The other thing is you undergo a lot of attacks today in politics. It happens to celebrities, but they tend to have at least some sort of fan base that are constantly supporting them. With politics today, you can often be in a situation where you're almost dehumanized. You're subject to attacks on your integrity, your character, your intentions.

It's possible, if you're not careful, that you just end up sitting there thinking this is really unfair. You get distracted from focusing on the business. That's why I always say to people, one part of being a political leader, or any leader, is to be able to have a certain sort of Zen. You need a bit of a Zen-like attitude to all the criticism and the disputatiousness that will go on around you. It's just going to happen.

Today with social media, it happens to an even greater degree. I often say to leaders that you cannot pay attention to this stuff. Get someone to summarize it for you in half a page and you read it in the morning.

Dwarkesh Patel 00:46:38

GPT can do it for you now.

Tony Blair 00:46:40

Honestly, if you start going down that rabbit hole, you'll never re-emerge.

00:46:45 – Looking back on the unipolar moment

Dwarkesh Patel 00:46:45

Let’s go back a little bit to your time in office. There was a unique unipolar moment that happens very rarely in history. The Anglosphere had much more power in the 1990s and early 2000s than the rest of the world. In what way did that feel different from today's world?

Is there something you would now change about the way the institutions were set up at the time and carried forward? There was a key opportunity in the unipolar moment. How should that have been used? How well was it used?

Tony Blair 00:47:19 

It's difficult. We did try a lot, contrary to what's sometimes written, for example with Russia. I dealt with President Putin a lot when I was prime minister. It was myself and President Clinton who took the crucial decision to bring China into the world's trading framework. The G7 at the time was the G8 with Russia there and China would always be invited.

I honestly think we tried a lot to recognize that we were going to live in a new world. The power wasn’t going to shift from the West in the sense the West would become not powerful. The East was going to become also powerful. We did understand that and work towards that.

Particularly in these last few years, China and Russia have come to a position that is seemingly hostile to Western democracy in terms of fundamental values and systems. That's a difficult problem. What we probably underestimated was how fast India would rise. At the time it seemed India was still going to be quite constrained.

We live in a multipolar world today. Personally, I think that's a good thing. In any event, it's an inevitable thing. It's really important always to give this message to China, for example. China as of right is one of the big powers in the world and as of right should have a huge influence. I don't believe in trying to constrain or contain China.

We do have to accept that the Chinese system is presently run on different lines to our own and is overtly hostile to some degree. This is why it's important for us to retain military and technological superiority, even though I believe passionately that it's important that we leave space for cooperation and engagement with China.

Now how much could we have foreseen all of this back in those days? I'm not sure. Sometimes one of the problems of the West is that we always see it through our own lens. We always think that we could have done something different to change the world. But actually the rest of the world operates on its own principles as well. Sometimes the change happens, not because we didn't do something, but because the rest of the world did.

Dwarkesh Patel 00:50:24

Final question. You interact with leaders today across dozens, maybe hundreds, of countries. Which among them is best playing the deck they've been dealt? Who is the most impressive leader? It doesn't have to be a huge country or anything. Given the deck they've been dealt, who's best?

Tony Blair 00:50:47

One of the things you must never, ever do in politics is say who's your favorite leader, who's done well. You will make one friend and many enemies. I'm just going to answer it in this way. If you look at the countries that have succeeded today — any country that has transitioned from being a third world or second world country to a first world country — certain things stand out and are clear.

Number one, they have stable macroeconomic policy.

Number two, they allow business and enterprise to flourish.

Number three, they have the rule of law.

Number four, they educate their people well.

Wherever you look around the world and you see those things in place, you will find success. Whenever you find their absence, you will find either the fact or the possibility of failure.

The one thing, however, that any country leader should focus on today is the possibility of all of these rules being rewritten by the importance of technology. The single most important thing today, if I were back in the frontline of politics, would be to engage with this revolution and to understand it. We need to bring people who also get it into the discussions and the councils of government and to take the key decisions that will allow us to access the opportunities and mitigate the risks.

Dwarkesh Patel 00:52:32

That's a wonderful place to close. Mr. Blair, thank you so much for coming on the podcast.

Tony Blair 00:52:35

Thank you for having me.