Hey, Vsauce. Michael here.
Do you want to be infected with Ebola without having to leave your own home or
deal with other people? Well, you might be in luck. You can
already download an Ebola virus genome. Right here on the Internet, right now.
And if you're willing to wait a few years for 3D bioprinting
technology to progress a little bit, you can just acquire one
then, submit the genome to it and ta da! All you can print Ebola. Or anthrax or whatever it is you wish to
mass-produce at home to wipe out humanity. Are humans going to go extinct soon? Will human extinction be anthropogenic? That is the result of human action. Or will it be one of the good old-fashioned kinds of extinction Earth's history knows pretty well? The Global Catastrophic Risks Survey,
issued by Oxford University's Future of Humanity Institute placed our risk of extinction before the year 2100 at 19%. Now, you might be thinking "whatever, blah
blah blah armageddon". "It'll be okay, humans are too smart to go extinct." Maybe you're right. But it's difficult to predict the distant future with a lot of certainty. What's really cool though is that if you embrace that uncertainty,
a simple argument can show that human extinction soon is actually more probable. It's called the Doomsday argument. Imagine a giant urn that contains either 10 balls numbered 1 to 10, or a million balls numbered 1 to a million. Now, you don't know which is the case, but you are allowed to pull out one ball. You go ahead and do that and it is ball number 4. That's pretty strong evidence in favour
of the 10 ball condition because drawing a four from a set of 1 through 10 is a one in 10 chance. But drawing four
from a million different numbers is a one in a million chance. By analogy you are also a numbered ball. You are a human who knows approximately what your birth number is. It's probably somewhere around 100 billion. That's how many other humans were most likely born before you were. Importantly, you didn't get to decide which birth number you would have. So, just like the number for a ball, you are a random sample from the set of all humans who will ever live. The Doomsday argument points out that from 200 billion people there's a
50 percent chance that a randomly chosen person, like you, would be born in the first one hundred billion. Whereas if there will be 10 trillion humans, there's only a one percent chance that any given human, say you, would happen to be born within the first 100 billion. Either you are special and lucky to be born so improbably
early in the story of humanity or your birth number is to be expected because there will not be tens of trillions of humans. Human extinction will be sooner rather than later. But before you become too convinced that the end is nigh, keep
in mind that the Doomsday argument is not uncontroversial. One problem it might have is a reference class problem. Are you really a random sample from the set of all humans who will ever be born? Well, if you believe that in the
not so distant future humans will be quite different than they are today. For instance, there'll be full of more 3D printed organs. The mere fact that right now
there aren't very many humans with that trait could be evidence that
you aren't a random sample from the set of all humans, just from the set of all humans like you, like does around you. Those born earlier in human history. Also the Doomsday argument
doesn't consider the likelihoods of actual threats or human advantages over those threats in the future. It just assumes that we don't know which way the balance will lie; that human extinction soon and human extinction later are equally likely.
But maybe you don't believe that. Maybe you are convinced that human ingenuity will always stay one step ahead of any extinction event thrown at it. You could be right, but there's reason to doubt that optimism. For example, the Fermi paradox. If it is likely that intelligent
life forms in our universe are capable of living for billions and billions of years, where are they? Why are the skies so silent? Perhaps it is because extinction level threat events are just too common for intelligent life anywhere to ever catch up. So, does this mean we should just give up? The Voluntary Human Extinction Movement think so. Founded in 1991, its supporters believe that humans are a negative influence on Earth and always will be. Thus we have a moral obligation to just stop reproducing right now and fade away. But what would a computer do? In a way, that's kind what Tom 7 did. He created a
program that plays video games. The program came up with novel
techniques and strategies for playing games and even exploited glitches humans didn't know about, or at least hadn't told it about. He also had the
program play other games, like Tetris, which I think is relevant to our question. The computer struggled to
figure out what to do. You see, the computer wasn't programmed
to consider future repercussions far enough ahead to notice that stacking Tetriminos in
certain ways made a big difference. On one run, when faced with imminent demise, the computer did something eerie. Rather than lose, and receive a 'game over', it just paused the game. For ever. Tom 7 describes the computer's reasoning like this: "The only winning move is to not play." And that's right. If you pause a game for ever you will never lose that game. But you'll also never win that game or achieve a high score. Now, we might not know what achieving a sentient life high score in this universe means or whether or not we're capable of
achieving one. We might also sometimes panic when the future looks bleak. But if we keep playing and keep learning, chances are we could eventually figure it out and start playing really well. So, thanks for continuing to play, for being here. And as always, thanks for watching.
Disappointed he didn't go into how much damage would have to be done to the earth to guarantee human extinction.
What fascinated me the most was the program that learned to play NES games and it's reasoning in pausing the Tetris game... it really put me into deep phylosofical tought. Truly interesting robotics stuff here.
5:30 -relavent xkcd http://xkcd.com/638/
I'm not sure the doomsday argument really suggests an extinction is likely to happen soon.
It merely suggests that human population will probably decrease significantly in the next centuries. It does not have to go all the way to zero, though.
missed the perfect chance to mention The Game
oh fun...