The Exponential Growth of Artificial Intelligence
In 2017, we can expect to see many self-driving cars on the road, more autonomy across many industries, and aggressive expansion of computer and robot assistance.
Many people have joked about the rise of Skynet (Terminator 2: Judgment Day) while others have expressed serious concern about the potential of such a thing happening. Many people including Elon Musk (SpaceX, Tesla, Open AI), Ray Kurzweil (computer scientist, inventor, futurist), Nick Bostrom (author of Super Intelligence), and Bill Gates (Microsoft) have shared their concerns about the serious threats we could face with creating a super intelligence.
So, what is a super intelligence and is it even going to happen?
As described by Bostrom in his book, a super intelligence is “an intellect that is much smarter than the best human brains in practically every field, including scientific creativity, general wisdom and social skills”. It will be a software created using deep computer learning algorithms with creativity capabilities, unlimited bandwidth, consciousness, and processing power which dwarfs the human mind.
Will it happen? Most certainly. When? That’s the million dollar question.
The opinion of scientists, enthusiasts, theorists, and enthusiasts range from decades to centuries. Jay Wheeler (author of 2050: Being Human in the Future) predicts it could arrive by 2030, due to the exponential growth of technology, which he explains in depth in his book.
An accurate answer as to when a super intelligence will be created is still unknown. So what’s the big deal about AI an an ASI (Artificial Super Intelligence)? If an ASI is created and we either control it or it is an ally, it would be able to process, learn, and test experiments a million times faster than the smartest human mind and it would be more intelligent (potentially in hours) than all of humanity combined. This would mean it could solve problems much faster than humans and give us answers to some of the biggest problems we face as a people. We’re talking, cure for cancer, political and economic solutions, and the answers than will allow us to travel at the speed of light and put humans on habitable planets beyond our solar system.
That sounds incredible, so what is all the worry about and what’s all this talk about rise of the machines?
An ASI would be more of a software than actual hardware, although it could easily be implemented into hardware, a robot’s “brain”, which is why many consider a ‘rise of the machines’ type scenario happening. Mostly though, it is a joke thrown around by people who have little understanding about an artificial super intelligence.
The worry is that a super intelligence would have incredible power and once uploaded to the internet (to learn), it could spread to every device connected to the internet, it could disguise itself, prevent itself from being turned off and take control of everything run on computers, which today is everything. Short of an EMP blast over the entire earth, it would be very difficult to shut down a rogue ASI. This would only be an issue if it became malicious. It could process all of humanity’s knowledge, review all our history, understand how we treat each other, and decide we need to be destroyed. There’s a million and one ways this could occur, but you can use your imagination of what it could do with control of everything from nuclear, water, gas, electric, resources, and everything else that runs on computers.
Linear growth, like how our brains work, is 1, 2, 3, 4… Where as exponential growth, doubled with each increase, is 1, 2, 4, 8. Although in this example there doesn’t seem to be much of a difference, after 30 increases, linear would obviously be at 30, while exponential growth increased would be a over a billion, yes, over 1,000,000,000. Here is a chart from Wheeler’s 2050: Being Human in the Future book to show the power of exponential growth and just why technology and super intelligence are becoming a extremely popular as of late:
As you can see from Wheeler’s example above, we appear to be approaching a point where everything is growing so rapidly, something is going to need to change in order to continue. Knowledge is doubling extremely quickly, but as mere humans we’re not able to process information faster than we could thousands of years ago.
We will need a super intelligence to process and manage all the information from the world or we will need to merge with machines.
More Technology Articles >>>