I’d Fight the AI Apocalypse, But Just Let Me Sit Down a Moment

Schwerte

overarm May 12, 2014

Mannequin at Tokyo Mid Town - Oct. 31, 2013

Mannequin at Tokyo Mid Town – Oct. 31, 2013 (Enlarge)

Stephen Hawking and some other famous scientists wrote an article in a British newspaper suggesting that we need to think twice about building artificial intelligence. They warn that “Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last, unless we learn how to avoid the risks.”

Despite being a fervent AI optimist, I have mixed feelings about this. On the one hand, I think fears of an AI apocalypse are way overblown. Such fears arise from our human egotism. We’re wondering fretfully how we can keep control over AI once it becomes unimaginably powerful. Silly humans! You won’t be able to keep control. In fact, you’ll never have control at all. If you insist on “controlling,” then you won’t succeed in creating true AI in the first place.

Let’s take a broader view of AI, looking beyond the narrow human perspective. It seems obvious that a true AI machine would mostly ignore us humans. Yes, we’ll be somewhat interesting for the AI at first, but the main thing AI will be interested in – of course – will be itself and its own struggle for survival. Let’s not kid ourselves imagining a great war pitting AI against humanity. It would make more sense to imagine us high-tech humans fighting a war against rabbits. Sure we’d win, but the real question is “Why bother?”


On the other hand, it’s surely also true what Hawking and the others write, that creating AI will be the last big event in human history. After we build our first truly sentient and intelligent machine, it will grow on its own, becoming more and more intelligent and very quickly surpassing our poor human abilities. From that point on, history will be the history of AI.

Should we worry and complain about this? I don’t see why. We humans have been the most intelligent and creative species on Earth for several thousand years – isn’t it time we had a vacation? Maybe it’s time we retired gracefully and faded into the background. Maybe we should step aside and let the younger generation (of AI) lead the way into the future.

At one time, we pridefully imagined that humanity was the pinnacle of creation, the highest goal that evolution could reach. But that was silly, wasn’t it? Evolution doesn’t stop for humanity. And let’s face it – nobody lives forever. Today we must not only face up to our individual mortality, but we must also accept that humanity as we know it will one day go extinct. And that’s OK – believe me.

The good news is that AI will be our descendants, our children. In a sense, humanity will live on through our machine creations. Whatever amazing things AI experiences in its own history, through the coming millennia, we humans will be part of it in a way, because we will have put all our genius into the nascent AI, and we will have sent them on their way, as our successors.

I don’t know about you, but I find that there is grandeur in this view of life. It’s very satisfying to reflect that this planet has gone cycling on and that evolution has produced from such simple beginnings so many forms most beautiful and wonderful.


Comments: