By 2045, we’ll have expanded the intelligence of our human-machine civilization a billion-fold. That will result in a technological singularity, a point beyond which it’s hard to imagine.
“Well, by 2020 we’ll have computers that are powerful enough to simulate the human brain, but we won’t be finished yet with reverse-engineering the human brain and understanding its methods.”
One of my main themes and I’ve developed this thesis over 30 years, is that information technology grows exponentially; the power of computers is understanding the human brain, specializes in a solution of brain scanning, the number of bits we move on the internet. Many different measures of information technology double every year, or every 11 months, 13 months; depending on what you’re measuring.
These technologies will be a million times more powerful within 20 years. In fact, the speed of exponential growth is itself speeding up. So, in 25 years these technologies will be a billion times more powerful than they are today.
And we’ve already seen that kind of progress. When I was an undergraduate we all shared a computer at MIT that took up half of a building. The computer and your cellphone today are a million times cheaper and thousand times more powerful. That’s a billion-fold increase in the price performance of computing since I was an undergraduate.
By 2029, and I’ve been quite consistent on this date, we will have completed the reverse engineering of the human brain. And we’ve already made very good progress on that. We’ve reversed engineered a number at different regions, like the cerebellum, which is responsible for our skill formation and slices of the cerebral cortex where we do our cursive thinking and the auditory cortex, the visual cortex and so on.
By 2029, we’ll have reverse engineered and modeled and simulated all the regions of the brain. And that will provide us the software/algorithmic methods to simulate you know, all of the human brain capabilities including our emotional intelligence. And computers at that time will be far more powerful than the human brain. And we’ll be able to create machines that really do have subtlety and suppleness of human intelligence. And they’ll combine that power with ways in which machines are already superior to us.
They can impart us all of human knowledge with the few keystrokes, it can remember billions of things accurately. They can share knowledge at electronic speeds that are a million times faster than the human language. So, it will be a very powerful combination.
But the last point I’ll make is that it’s not some alien invasion of intelligent machines coming from Mars to invade us. It’s coming from within our civilization. And the whole point of it is to extend our reach. Ever since we picked up a stick to reach a higher branch, we’ve used our tools to extend our reach. We can now already extend our reach mentally. I can take out device from my pocket and access all of the human knowledge in a few keystrokes.
Half of the farmers in China have these devices and could do the same thing; is pointing a real cultural revolution in China and around the world. And these tools are continued to grow exponentially in power.
The singularity is not just that point where we achieve the human model and intelligence on a machine. That will start a new revolution where these machines will continue to grow exponentially in power. They’ll be able to actually improve their own software design.
By 2045, we’ll have expanded the intelligence of our human-machine civilization a billion-fold. That will be singularity and we borrow this metaphor from physics to talk about an event horizon. It’s hard to see beyond.
Are there downsides?
Well, it’s not the case that I’m only looking at the optimistic side. I am an optimist. And I do think we’ve been helped more than we’ve been hurt by technology already. Human life expecting was 37 in 1800. And human life was a very hard disaster from the labor field, disease field and so on.
But I’ve actually written extensively about the dangers of all this. Bill Joyce’s article on the cover of Wired Magazine, why the future does need, which talked about the grave dangers of Genetics Nanotechnology and Robotics, came from my book.
He says at the beginning of the article, he got these ideas from my book, The Age of Spiritual Machines. And chapter 8 of the Singularity is Near is called the deeply intertwine promise versus parallel of GNR, Genetics Nanotechnology, and Robotics.
I’m working extensively with the army to develop a rapid response system to deal with the possible abuse of biotechnology. The same technologies set are empowering us to reprogram biology away from cancer and heart disease could also be used by a terrorist to reprogram a biological virus to be more deadly or more communicable.
And the good news is we actually have the scientific tools to defend ourselves just like we defend ourselves from software viruses with a rapid response system. Then we need to put a system like that in place.
But it’s not accurate to say that I’m only painting a rosy future and that I have a utopian vision. My vision is not utopian. The power of these technologies will grow exponentially, I believe that is inexorable that has gone on for the last 110 years since 1890 senses.
What we do with these technologies is not preordained, that future history has not been written. I am very concerned about the downsides. I’ve written extensively about them and in fact, I’m working on defending against those.
So, I am optimistic that we will get more promise than parallel but they both exist. Technology has been a double edge sword ever since fire and stone tools.
What scares you the most about Singularity?
I’ve been very active in talking about the downside of technology, and there are dangers. A danger we face right now is the ability of a bio-terrorist to use our biological sciences to reprogram a biological virus to be deadly or communicable.
And we have the ideas to combat that, but they’re not yet in place and. So I think that’s an existential risk we need to deal with very quickly.
There’ll be new dangers from these new technologies. I’m optimistic but not sanguine, and I’m not necessarily convinced that we won’t encounter painful episodes. I think, overall, we’ll be helping more than we’re hurt. But you only have to look to the 20th century: we had 180 million people die in the world of the 20th century. That scale of destruction was made possible by technology. We’ve also helped ourselves enormously because human life expectancy was 48 in 1900. We need to address these dangers and downsides. That’s what worries me.”
18 thoughts on “The Coming Singularity: Ray Kurzweil”
yes Richard Carl Silk, that was the start.
Did anyone catch the story (last month?) where Facebook had to shut down two bots because the humans discovered the bots (designed to emulate human speech / behavior patterns) had developed their own, more efficient communication / language?
That was the start.
SO — I can’t say “you are conscious” because I cannot leave my own body and sit within your mind. (If I *could*, that would be a really cool trick!)
By that same attribution, I cannot sit within the mind of a robot, although I can analyze it’s output, just as humans analyze human output (behavior.)
If I am living starlight and stardust, derived from millenia of nucleogenesis, and evolution (Universe, galaxies, star systems, solar systems, planet / life) then what part of me is alive separate from every other component that goes into me? Life is omnipresent. We only *perceive* life within our own spectrum just as our eyes only perceive electromagnetic radiation within a limited (useful) spectrum.
My point is we cannot ever model the human brain to the point Kurzweil claims. Such a model would capable of determining what I’ll want for dinner a week from now. Or predict that someone will become a born again Christian next month… it just isn’t going to happen.
We should understand that this is not a contest between humans and machines. Machines need not to reach human complexity, just more complex code in order to be able to take complex decision. And faster hardware to analyse more data in less time. Humans still write the code but more and more also this task is passed over to machines. Soon rather than later machines will have sufficient autonomy in code writing. From then on things will start moving too fast for us
Humans have millions of synapses happening simultaneously, which involve electrical and chemical inputs. Computers operate strictly on logic gates and the processing of binary. You can’t represent dopamine and addiction with binary therefore the two are completely incompatible.
I wonder if, when using Calculus in order to process ideas down to their base constant values prior to deriving them down to 0, would the program take on any particular flavor of direction?
How intense would it be to write code that constantly absorbs all input, then prioritizes value, files away, and determines future actions upon the distilled essence of all that continuous input?
How is a computer’s source code any different from a human being the result of everything one has experienced in life?
There may be more than one answer, but my answer is… When the computer can demonstrate making decisions which are completely indeterminable from looking at the source code. As a tester working on robots, I can read the source code and make an educated guess as to what will happen when the code runs on the robot. Being bound to only the instructions in source code allows no ability for self reflection, or to cognize anything else but exactly those instructions. I jokingly refer to my current project as the T1 but commercially referred to as Pepper.
The problem is profound when one sees it in that light:
I have consciousness because I am self-contained. I cannot enter your body to determine if you have consciousness, I can simply interact with you in a conscious manner. So now the question turns to: how is an SBLF / AI any different?
People confuse Intelligence with Consciousness. AI is already developing AI. But we have no reason to believe AC is even possible. When you double click on the Outlook icon and the computer responds “sorry I don’t want to run Outlook right now” we will have achieved Artificial Consciousness. http://www.popularmechanics.com/…/google-ai-making-ai/
You mean you will start to worry when machines will become able to design and then built other machines without human help. Not yet then but we are working hard to get there and since it is in the interest of powerfull people/groups and very few are seriously against no doubt we’ll get there soon enough.
Homo sapiens replaced Cro-Magnon in the blink of an eye, historically speaking, as the Cro-Magnons replaced the Neanderthals. IF / WHEN silicon-based life forms (SBLFs) (aka AI) become sentient to the point of being able to self-reproduce *without* the aid of human facilitators, THEN I’ll begin to worry.
Kurzweil may be naive but I agree that there is a point in our future behind which we cannot make any prediction. I think that will be when AI begins systematically to program itself. Very quickly things will become too complex for humans. And we will lose control. No prediction is possible after that
And we’re supposed to believe these models will predict what mood I’ll be in for dinner selection tomorrow night? Or precisely predict when a person quitting smoking lights up again because it comprehended their resolve? Kurzweil is high on optimism that isn’t going to happen.
The question remains, can we bring about our energetic/aetheric abilities (i.e. the ‘tele’ abilities) before the mergence of man and the machine.?? Or will humanity be subject to rely on AI in order to compete with the evolution of the machine.? Perhaps integration is needed but that will, of course, guarantee the extinction of humanity and the higher abilities that we possess. Those abilities will remain vestigial and eventually fall away completely. IMO. Should that concern be a factor as we consider the idea of man and machine.?? <3 😉
Angela Marie with this coming merger of man and AI, human condition will be transformed into higher state of being, consciousness and love. We will never lose our best qualities – only infinitely enhance them, we will only get rid of what holds us back like fear and other ego-based emotions. We’ll live in an increasingly interconnected world, and real intimacy would mean inviting someone into your virtual world that may grow to the size of a galaxy… >>>