Previous: 7 Ways to Spruce Up Your Cooking with Science
Next: Cybersecurity & a Hacker Called Alien | SciShow Talk Show



View count:301,730
Last sync:2022-11-25 19:00
Go to to start streaming The Joy of Chance. Use the promo code ‘scishow’ during the sign-up process to get your first 30 days free.

We humans like to think we’re special in basically all ways, but if the history of life is any indication, our species has a limited time on this planet. So the question is: when are we gonna go extinct?

Hosted by: Hank Green

SciShow has a spinoff podcast! It's called SciShow Tangents. Check it out at
Support SciShow by becoming a patron on Patreon:
Huge thanks go to the following Patreon supporters for helping us keep SciShow free for everyone forever:

Adam Brainard, Greg, Alex Hackman, Sam Lutfi, D.A. Noe, الخليفي سلطان, Piya Shedden, KatieMarie Magnone, Scott Satovsky Jr, Charles Southerland, Patrick D. Ashmore, charles george, Kevin Bealer, Chris Peters
Looking for SciShow elsewhere on the internet?

Image Sources:
Thanks to CuriosityStream for supporting this episode of SciShow!

Go to to learn more. We humans like to think we're special in basically all ways, but if the history of life is any indication, our species has a limited time on this planet and in this universe.

So the question is: when are we gonna go extinct? To answer this, you might start to think about the odds of a large meteor impact, or a nuclear war. But you can actually make a pretty decent guesstimate just by knowing how long we've already been around.

You don't need to know the likelihood of different extinction events or anything about human behavior specifically— just some clever statistical tricks. The mathematical methods for working out the timing of humanity's final curtain call have been given the spooky name of the Doomsday Argument. There's nothing spooky about it.

It's actually based on a concept from cosmology called the Copernican Principle. Copernicus was the guy who proposed that the Earth was just one planet of many revolving around the Sun — an idea that was pretty radical at the time. And since then, every time we've learned more about the cosmos, we've found that our place in it is less special and unique than we used to think.

In the spirit of Copernicus, cosmologists have made this idea a fundamental principle: your default assumption should be that you are an average observer. You are not special! And in a 1993 Nature paper, astrophysicist J.

Richard Gott applied this idea to our place in time as well to arrive at a rough idea of when our species will die out. It turns out all you need to know is how long our species has been around already. According to ancient DNA evidence, the species Homo sapiens has been around for about 350,000 years.

And under the Copernican principle, we should assume that now isn't a special period in humanity's total lifespan. That makes sense—if you think about it, it would be pretty ‘special' if you, the person watching this video, happened to be living during the first or last 1% of humanity's existence. But if you are not a special observer, then you're equally as likely to be at any point from the start to the end of the human race.

Viewing that from a probabilistic lens means you can be 95% confident that now represents a time somewhere in the middle— between 2.5% and 97.5% of the entirety of human existence. That's what statisticians call a confidence interval. So if you crunch the numbers assuming that 350,000 years is either 2.5% or 97.5% of the total timeline, you get that we've got between about 9,000 and 13.7 million years.

That's a pretty big range, and while I'm glad to hear we've probably got at least 9000 years left,. I would like to know things with a little more certainty than that! But unfortunately that's the best this argument can give us.

Narrowing the range means lessening our confidence. And there's already a 5% chance that we're wrong—and humanity will disappear much sooner or long after that interval. Of course, this is not the only statistical way to look at human existence.

Another variant of the Doomsday Argument can be found in a 2016 paper by astronomer Fergus. Simpson. This paper comes at the question from a different angle.

Rather than looking at time, he thought about the total number of people who ever will live. And because we're talking about population sizes, he argued that the probability for the total number of people who will ever live should follow what's called a power law distribution, where likelihood decreases with size. This was based on research into the size distribution of things like cities and microbe populations.

This is complicated statistics stuff, so let's hear how philosopher Nick Bostrom explained the argument. Imagine you have a jar with some number of numbered balls in it, but you have no idea of how many balls are in there, you can't even look inside. But you do know that the balls are numbered 1, 2, 3 and so on.

If you draw out the number 7, you can ask yourself which is more likely, that the jar contains 10 balls, or a million balls? The answer of course is 10. Drawing a ‘7' is much more likely if there are 10 balls than if there are a million balls in which case you'd probably get some 6 digit number printed very small on the marble.

So by knowing how many humans have lived so far, by knowing like what your number is, if you were one of those balls in a jar, you can guess how many will live in the future, thanks to some complicated statistical equations. Researchers estimate that about a hundred billion human births have occurred to date. And when you plug that into the statistical formulas, the future looks pretty bleak.

Assuming the birth rate doesn't change, Simpson predicted that in the 21st century, there's about a one in 500 chance of human extinction every year. That means that the total chance of extinction before 2100 is over 10%. And the odds that we'll make it to the year 3500?

That's only 40%, or 2 in 5. These two researchers got very different answers for their Doomsday Arguments, and I like the first one more than the second one. If you Google around, you can find other variations of this argument that give us more different results.

But it's all math, so what gives? Well, a statistical model is only as good as your data and your assumptions. On the data side, one argument relies on the time humanity has been around; the other relies on the number of people who have ever lived and the current birth rate.

Because so many more humans are being born now than in the past, the end becomes much closer in the second argument. And on the assumptions side, while everyone agrees that we should assume we're ‘typical' observers, defining what ‘typical' means is really tough when you only have one data point! These are the kinds of things people point out when they criticize the Doomsday Argument.

While most agree with the basic premises of the idea, specific versions of the theory, like the ones mentioned here, tend to have flaws in their exact methods. Like, when is the start of humanity? The appearance of Homo sapiens?

The start of civilization? The first time any of our ancestors used fire? This is tricky stuff!

Also, don't forget that the Copernican principle assumes you have no good reason to think you're special. If you can see a giant meteor about to hit the Earth, then you have a pretty good reason to assume you are in fact a special observer — one of the last humans ever. Sorry!

But just because it's hard to figure all this out doesn't mean it's not worth the exercise. The Doomsday Argument is maybe a little bit morbid, but thinking about it is also kind of important. If we find out that the risk of a global catastrophic event is 0.2% per year, then we can make informed decisions about how to spend resources on defenses against those risks.

Plus, it forces us to ask one of the biggest questions imaginable: what do we want the future of humanity to look like? And how do we get there? If you like thinking about the big questions in life -- and of course you do, because you're watching SciShow -- we think you might enjoy the videos offered over at CuriosityStream.

Today's episode is brought to you by CuriosityStream, which is a subscription streaming service that offers over 2400 documentaries and non­fiction titles from some of the world's best filmmakers, including exclusive originals. They have videos on nature, history, technology — even society and lifestyles — which is one of the reasons we like them so much. Like, if you liked this episode, you might enjoy The Joy of Chance.

It's a whole documentary that dives into the role uncertainty in our universe. You can get unlimited access to content like this starting at $2.99 a month. And as a special thanks to our SciShow audience,you can get the first 30 days for free!

You just have to sign up at and use the promo code “scishow” during the sign-up process. {♪OUTRO}.