It's not always as straightforward as it may seem to measure the size of a city or urban region.

There's the problem of which urban boundary to use. And then once you've landed on that, there's the additional problem of estimating how many people live within it. This can be particularly challenging when it comes to informal settlements, where there isn't reliable population data.

The most common approach is to use the continuous built-up area as the urban agglomeration, as opposed to any sort of "city proper" boundaries. And this is exactly what the United Nations has done in its latest World Urbanization Prospects report.

Here, they have fundamentally revised their measurement methodology by using a new, so-called harmonized geospatial approach. If you'd like to nerd out on the specifics, you can do that here.

But one of the key takeaways is that this new methodology has resulted in a reordering of the world's largest urban agglomerations. At the top is now Jakarta, followed by Dhaka:

Previously, Tokyo was thought to be the world's most populous megacity, but it has dropped down to 3rd in this new report. And by 2050, it is forecasted to drop even further to 7th place:

As we spoke about yesterday, the world's economic center of gravity is rapidly shifting toward Asia. And that shows up in these charts.

I have a strong desire to visit the largest cities in the world. It's fascinating to see how such large urban clusters manage to organize themselves. There are always systems that naturally emerge to make things work, even if it feels chaotic on the surface.

I've only been to 3 cities from the 2025 chart, so I have my work cut out for me.

Cover photo by Rifki Kurniawan on Unsplash

I have a new not-so-serious goal. I say not-so-serious because I'm busy with lots of other things and who knows if/when I'll actually get around to this. So let's just say that this goal is based on a strong interest. I would like to visit the 10 largest urban areas in the world.

I was careful to say urban areas, because largest cities can be misleading. City proper boundaries are an arbitrary measure. What matters more are the built up areas.

I have already been to a handful of them, namely the ones centered around Tokyo, Dhaka, and Mexico City. But there are more on the list that I haven't been to.

There are, however, two immediate challenges with setting a not-so-serious goal like this one. The first is that it's not easy to come up with a definitive list of the largest urban areas. It's nuanced and, for some cities, population estimates are very much rough estimates.

The second is that population figures are constantly changing. So by the time I get around to this, many of the largest cities might be in Africa (current forecast for this century) and I may be checking off the wrong list.

But let's not let perfect be the enemy of good. Here's what I came up with. And by "what I came up with," I mean that I asked ChatGPT its opinion. This is the list I got:

Then I asked the question in a slightly different way and got this list:

Then I asked it to tell me its definition of metropolitan area vs. urban agglomeration:

Finally, I asked it why Lagos, Nigeria was not on the list. Current estimates place this urban region at more than 20 million people (which would place it ahead of New York in the first list), but I don't think anyone really knows for sure. Whatever the current number, it is widely understood to be one of the fastest growing city regions in the world. This is how ChatGPT responded:

I'm a little torn because high on my list of cities to visit are São Paulo and Seoul. And they don't seem to be finding themselves on the same list. I'm also very curious to see Lagos, and I have a suspicion that it's much larger than official estimates. Regardless, there are some clearly big cities to check off. India and China would be good places to start. So I better get to work. Or not. Since this is a not-so-serious goal.

Toronto has a lot more CCTV cameras than I would have thought.

According to this (2022?) data from Comparitech, there is estimated to be about 19,236 cameras installed around the Greater Toronto Area. With a population of around 6.31 million people, this translates into a per capita rate of 3.05 (CCTV cameras per 1,000 people). What this means is that there is almost surely footage of me enjoying a late-night shawarma sandwich after the bar somewhere on the streets of Toronto.

In some ways, this is a high number of cameras. Tokyo, which is usually considered to be the largest metro area in the world with nearly 40 million people, only has 1.06 cameras per 1,000 people. Dhaka is 0.71. Sao Paulo is 1.04. Osaka is 1.57. And Montreal is 1.03. Though to be totally fair here, Rio de Janeiro is up at 3.34 (and it may be the most dangerous city mentioned in this post). Paris is 4.04. New York is 6.87. Los Angeles is 8.77. And London is 13.35.

But where things get really exciting is in authoritarian places. Moscow is estimated to have 16.85 CCTV cameras per 1,000 people. And in China as a whole, there is estimated to be roughly 540 million cameras scattered around its cities, which works out to an average of 372.8 cameras for every 1,000 people. For a city like Shanghai, this crudely equals something like 10.6 million cameras.

It turns out that surveillance is pretty important for things other than shawarma-eating videos:

Vyborov wasn’t arrested that day, but the police informed him that he was under surveillance through Sfera, one of Moscow’s face recognition systems, for participating in unsanctioned rallies. Considered one of the most efficient surveillance systems, Sfera led to the detention of 141 people last year. “Facial recognition, and video cameras in general in a totalitarian state, are an absolute evil,” Vyborov says.

Here's the other thing. Safety is usually touted as the reason to have lots of cameras. But Comparitech's data suggests that there's an almost non-existent correlation between lots of cameras and lower crime. I mean, just look at Tokyo. It is basically the model megacity, and its per capita camera rate is only 1.06. The real utility, it would seem, is using cameras and face recognition software to restrict personal freedoms.

It's not always as straightforward as it may seem to measure the size of a city or urban region.

There's the problem of which urban boundary to use. And then once you've landed on that, there's the additional problem of estimating how many people live within it. This can be particularly challenging when it comes to informal settlements, where there isn't reliable population data.

The most common approach is to use the continuous built-up area as the urban agglomeration, as opposed to any sort of "city proper" boundaries. And this is exactly what the United Nations has done in its latest World Urbanization Prospects report.

Here, they have fundamentally revised their measurement methodology by using a new, so-called harmonized geospatial approach. If you'd like to nerd out on the specifics, you can do that here.

But one of the key takeaways is that this new methodology has resulted in a reordering of the world's largest urban agglomerations. At the top is now Jakarta, followed by Dhaka:

Previously, Tokyo was thought to be the world's most populous megacity, but it has dropped down to 3rd in this new report. And by 2050, it is forecasted to drop even further to 7th place:

As we spoke about yesterday, the world's economic center of gravity is rapidly shifting toward Asia. And that shows up in these charts.

I have a strong desire to visit the largest cities in the world. It's fascinating to see how such large urban clusters manage to organize themselves. There are always systems that naturally emerge to make things work, even if it feels chaotic on the surface.

I've only been to 3 cities from the 2025 chart, so I have my work cut out for me.

Cover photo by Rifki Kurniawan on Unsplash

I have a new not-so-serious goal. I say not-so-serious because I'm busy with lots of other things and who knows if/when I'll actually get around to this. So let's just say that this goal is based on a strong interest. I would like to visit the 10 largest urban areas in the world.

I was careful to say urban areas, because largest cities can be misleading. City proper boundaries are an arbitrary measure. What matters more are the built up areas.

I have already been to a handful of them, namely the ones centered around Tokyo, Dhaka, and Mexico City. But there are more on the list that I haven't been to.

There are, however, two immediate challenges with setting a not-so-serious goal like this one. The first is that it's not easy to come up with a definitive list of the largest urban areas. It's nuanced and, for some cities, population estimates are very much rough estimates.

The second is that population figures are constantly changing. So by the time I get around to this, many of the largest cities might be in Africa (current forecast for this century) and I may be checking off the wrong list.

But let's not let perfect be the enemy of good. Here's what I came up with. And by "what I came up with," I mean that I asked ChatGPT its opinion. This is the list I got:

Then I asked the question in a slightly different way and got this list:

Then I asked it to tell me its definition of metropolitan area vs. urban agglomeration:

Finally, I asked it why Lagos, Nigeria was not on the list. Current estimates place this urban region at more than 20 million people (which would place it ahead of New York in the first list), but I don't think anyone really knows for sure. Whatever the current number, it is widely understood to be one of the fastest growing city regions in the world. This is how ChatGPT responded:

I'm a little torn because high on my list of cities to visit are São Paulo and Seoul. And they don't seem to be finding themselves on the same list. I'm also very curious to see Lagos, and I have a suspicion that it's much larger than official estimates. Regardless, there are some clearly big cities to check off. India and China would be good places to start. So I better get to work. Or not. Since this is a not-so-serious goal.

Toronto has a lot more CCTV cameras than I would have thought.

According to this (2022?) data from Comparitech, there is estimated to be about 19,236 cameras installed around the Greater Toronto Area. With a population of around 6.31 million people, this translates into a per capita rate of 3.05 (CCTV cameras per 1,000 people). What this means is that there is almost surely footage of me enjoying a late-night shawarma sandwich after the bar somewhere on the streets of Toronto.

In some ways, this is a high number of cameras. Tokyo, which is usually considered to be the largest metro area in the world with nearly 40 million people, only has 1.06 cameras per 1,000 people. Dhaka is 0.71. Sao Paulo is 1.04. Osaka is 1.57. And Montreal is 1.03. Though to be totally fair here, Rio de Janeiro is up at 3.34 (and it may be the most dangerous city mentioned in this post). Paris is 4.04. New York is 6.87. Los Angeles is 8.77. And London is 13.35.

But where things get really exciting is in authoritarian places. Moscow is estimated to have 16.85 CCTV cameras per 1,000 people. And in China as a whole, there is estimated to be roughly 540 million cameras scattered around its cities, which works out to an average of 372.8 cameras for every 1,000 people. For a city like Shanghai, this crudely equals something like 10.6 million cameras.

It turns out that surveillance is pretty important for things other than shawarma-eating videos:

Vyborov wasn’t arrested that day, but the police informed him that he was under surveillance through Sfera, one of Moscow’s face recognition systems, for participating in unsanctioned rallies. Considered one of the most efficient surveillance systems, Sfera led to the detention of 141 people last year. “Facial recognition, and video cameras in general in a totalitarian state, are an absolute evil,” Vyborov says.

Here's the other thing. Safety is usually touted as the reason to have lots of cameras. But Comparitech's data suggests that there's an almost non-existent correlation between lots of cameras and lower crime. I mean, just look at Tokyo. It is basically the model megacity, and its per capita camera rate is only 1.06. The real utility, it would seem, is using cameras and face recognition software to restrict personal freedoms.