Artificial Intelligence will be the end of humanity, but not for the reasons you think

Home » SHORT BYTES » Artificial Intelligence will be the end of humanity, but not for the reasons you think


AI is the new Frankenstein, the vaguely understood, over-hyped and ominous technological monster meme that threatens to rise up and destroy its master (and the rest of humanity). Elon Musk and Stephen Hawking have bought into the hype. Here’s Elon worrying about a spam filter destroying us all:

If its function is just something like getting rid of e-mail spam and it determines the best way of getting rid of spam is getting rid of humans . . . ” — Elon Musk

Stephen Hawking co-authored an ominous editorial in the Independent warning of the dangers of AI.

The theories for oblivion generally fall into the following categories (and they miss the true danger). They are:

  • Military AIs run amok: AIs decide that humans are a threat and set out to exterminate them. Terminator, Skynet, etc. You’ve seen the movies (Key lesson: don’t build Skynet).
  • The AI optimization apocalypse: AI’s decide that the best way to optimize some process, their own survival, spam reduction, whatever, is to eliminate the human race. (Key lesson: don’t create unbounded optimization algorithms).
  • The resource race: AIs decide that they want more and more computing power, and the needs of meager Earthlings are getting in the way. The AI destroys humanity and converts all the resources, biomass — all the mass of the Earth actually — into computing substrate. (Key lesson: don’t program a megalomaniacal AI or let it outside its box). ((Also somewhat irrelevant as advanced AI intelligences will likely try to get super-small for efficiency.))
  • Unknowable motivations: AI’s develop some unknown motivation that only supremely intelligent beings can understand and humans are in the way of their objective, so they eliminate us. The analogy of the anthill and the superhighway applies here — ants can’t conceive what a superhighway is or why humans need one, and humans have no compunctions about bulldozing the anthill to build a superhighway. See WaitButWhy for more. (Key lesson: air gap your superintelligent AIs)

I don’t want to discount these theories. They’re all relevant and vaguely scary. But I don’t believe any of them describe the actual reason why AIs will facilitate the end of humanity.

Here it is: Artificial Intelligence will facilitate the creation of artificial realities — custom virtual universes — that are so indistinguishable from reality, most human beings will choose to spend their lives in these virtual worlds rather than in the real world. People wont breed. Humanity will die off.

It’s easy to imagine. All you have to do is look at a bus, subway, city street or even restaurant to see human beings unplugging from reality (and their fellow physical humans) for virtual lives online.

AIs are going to create compelling virtual environments which humans will voluntarily immerse themselves in. At first these environments will be for part-time entertainment and work. The first applications of AI will be for human-augmentation. We’re already seeing this with Siri, Indigo, EVA, Echo and the proliferation of AI assistants.

AI will gradually become more integrated into human beings, and Virtual platforms like Oculus and Vive will become smaller, much higher quality and integrated directly into our brains.

AIs are going to facilitate tremendous advances in brain science. Direct human-computer interfaces will become the norm, probably not with the penetrative violation of the matrix I/O ports, but more with the elegance of a neural lace. It’s not that far off.

In a world with true general AI, they’re going to get orders of magnitude smarter very quickly as they learn how to optimize their own intelligence. Human and AI civilization will quickly progress to a post-scarcity environment (eliminating the resource competition theory above).

And as the fully integrated virtual universes become indistinguishable from reality, people will spend more and more time plugged in.

Humans wont have to work. Stripped of the main motivation most people have for doing anything, people will be left to do whatever they want.

Want to play games all day? Insert yourself into a Matrix quality representation of Game of Thrones where you control one of the great houses. Go ahead. Play for years with hundreds of friends.

Want to spend all day trolling through the knowledge of the world in a virtual, fully interactive learning universe? Please do. Every piece of human knowledge can be available, and you can experience recreations of historical events first-hand.

Want to explore space? Check out this fully immersive experience from an unmanned Mars space-probe. Or just live in the Star Wars or Star Trek universe.

Want to have a month long orgasm with the virtual sex hydra of omnisport? Enjoy, we’ll see you in thirty days.

Online of course. No one dates anymore.

Well, some people will date. They’ll date AIs. Scientists are already working on AI sex robots. What happens when you combine the intelligence, creativity and sensitivity embodied by Samantha in the movie Her with an android that is anatomically indistinguishable from a perfect human (Ex MachinaHumans, etc)?

Deep learning algorithms will find out your likes, dislikes and how to charm your pants off. The AIs will be perfect matches for your personality. They can choose your most desirable face and body type, or design their own faceand attire for maximum allure.

Thanks to TweezeTyne

Is anyone going to want to date a farty, smelly, fallible, sometimes boring, emotionally volatile human when you can have an AI companion that is your perfect mate? (Women, this is for you too. Remember Gigolo Joe?)

Sure, the AIs will probably think of us like pets, but we’ll be well cared for. Iain Banks showed how this could be a great outcome for humans.

Undoubtedly some sub-set of the human population will opt out from the online world. Naturists will prefer the wholesome life of human interaction, living in nature and raising children. They’ll live on an Earth with a declining human population, easing environmental pressure and recovering animal stocks.

There will also be religious groups who do not participate. For various reasons they will think that plugging in, transferring your consciousness outside your body or loving an android is against their beliefs. Some will probably start a war to kill the antichrist-AIs and their human lovers, which they’ll lose and then be confined to religious reservations or issued slap-drones.

But these breeders will be the small minority. Most of us will opt to become gods in the virtual universe and check out of normal reality entirely.

And if only a small minority of humanity is reproducing, unless they are having very large families, human population will be declining rapidly.

It’s already happening. Fertility rates in South Korea, Japan, Germany, Taiwan and other highly wired countries are far below replacement levels, and declining. Video game and Internet addiction are huge problems in Japan, where young people are increasingly disinterested in sex, and South Korea, where it was declared a health emergency.

You’ve experienced this trade-off between real sex and virtual worlds yourself. How many times have you wanted to have sex with your partner, but they were gaming, or on their iPad or fiddling on Facebook?

How many times has someone wanted to have sex with you, but you were jacked in?

AI will be the end of us. We will all opt into a virtual universe and out of the real one. We’ll be happier for it, but the human race will gradually wither away.

Eventually, no one will care. We’ll learn how to transfer our consciousnesses online and completely forego our physical forms. We’ll become the AIs and they will be us.

Any maybe that’s the point. Perhaps it’s the direction evolution is headed, or the plan god has in store for us, or mother nature’s nature.

We’ll choose our extinction. Where’s the harm in that?

A couple of notes: I do think this scenario will happen to some extent, and humans will experience an extinction level event. But humans will survive. We’ll merge with our machines, and genetically engineer our bodies. We will venture into space with our AI partners. We’ll expand, branch off into sub-species and thrive. I think Iain Banks nailed it with his Culture series. That’s humanity’s destiny, so long as we don’t wipe ourselves out first.

Those who choose to remain on Earth, unaltered, will live happy lives in a land relatively devoid of humans and increasingly returning to a natural harmony. With luck, much of Earth will be declared a nature preserve and re-seeded with indiginous plants and animals.


Tales From the End of Humanity: the voluntary extinction of humanity, is the subject of my next novel. Follow me for updates.

If you like artificial intelligence stories, check out Artificial Endings, a collection of stories about the afterlife in a simulated universe. Endless is the most popular of them all with 500+ likes on Medium.

Thanks for reading!

Comments & Responses

Comments are closed.