All Posts in Futurism

September 24, 2016Your thoughts?

What does the future look like when machines take over?

The hope is that they are ethical, and that they follow Asimov's laws of robotics and are kind to humans. Hope aside, let's think about what might be the real implication of all humanity allowing machines and artificial intelligence take over all rote and repetitive menial tasks.

The purpose of one's life on earth—take it for what it's worth—is the survival of the species. But there's a catch to this. It's the survival of the species that is best suited to survive. Is it a cybernetic species, or is it a sentient artificially intelligent species? Is it super-intelligence?

There's been a long drawn debate around direct correlation between intelligence and morality. One school of thought has maintained that the higher the intelligence, the higher the moral values of the entity possessing the intelligence. However, not only has US researcher Howard Gardner suggested that correlation between intelligence and morality is highly unlikely, but you can totally assume machines not coming anywhere close to being moral with the increase in intelligence, real or artificial.

You can totally assume Artificial Intelligence not coming anywhere close to being moral. Click To Tweet

While everybody is thinking AI would mean self driving cars eventually, we may still be surprised how soon abilities assumed by AI will tremendously outperform any human. How long do you think it will take until translation by machines will be better than anything humans can do? Same with Math.

Think about it. Look at the scale of progress around us. Most of scientific and technical progress that we have made as a species is just in last 200 years or so, in 3.8 Billion years of life on earth, and about 2.5 million years of humanity.

As Nick Bostrom says in his book, it might only take less than a month for AI to go from Village-Idiot-level General Intelligence (AGI) to Einstein-Level-and-Beyond Super Intelligence (ASI).

PPTExponentialGrowthof Computing

What do you think humans will be able to call their own?

Picture this. You're having salad, and a friend asks how the salad is. You answer "Not bad". The meaning of something being not bad in that context means so many things at the same time. It means good, okay, bearable, fine, I'm English (Queen's English) and many things in between. The day when AI picks up the tab on your salad, you're done.

Big time.

And here, done means 'finished' not completed. ■