We're also literally replacing sections of the brain that are degraded or don't work any more because of disabilities or disease. There are neural implants for Parkinson's Disease and well-known cochlear implants for deafness. There's a new generation of those that are coming out now that provide a thousand points of frequency resolution and will allow deaf people to hear music for the first time. The Parkinson's implant actually replaces the cortical neurons themselves that are destroyed by that disease. So we've shown that it's feasible to understand regions of the human brain, and reimplement those regions in conventional electronics computation that will actually interact with the brain and perform those functions.

If you follow this work and work out the mathematics of it. It's a conservative scenario to say that within 30 years — possibly much sooner — we will have a complete map of the human brain, we will have complete mathematical models of how each region works, and we will be able to re-implement the methods of the human brain, which are quite different than many of the methods used in contemporary artificial intelligence.

But these are actually similar to methods that I use in my own field — pattern recognition — which is the fundamental capability of the human brain. We can't think fast enough to logically analyze situations very quickly, so we rely on our powers of pattern recognition. Within 30 years we'll be able to create non-biological intelligence that's comparable to human intelligence. Just like a biological system, we'll have to provide it an education, but here we can bring to bear some of the advantages of machine intelligence: Machines are much faster, and much more accurate. A thousand-dollar computer can remember billions of things accurately — we're hard-pressed to remember a handful of phone numbers.

Once they learn something, machines can also share their knowledge with other machines. We don't have quick downloading ports at the level of our intra-neuronal connection patterns and our concentrations of neurotransmitters, so we can't just download knowledge. I can't just take my knowledge of French and download it to you, but machines can. So we can educate machines through a process that can be hundreds or thousands of times faster than the comparable process in humans. It can provide a 20-year education to a human-level machine in maybe a few weeks or a few days and then these machines can share their knowledge.

The primary implication of all this will be to enhance our own human intelligence. We're going to be putting these machines inside our own brains. We're starting to do that now with people who have severe medical problems and disabilities, but ultimately we'll all be doing this. Without surgery, we'll be able to introduce calculating machines into the blood stream that will be able to pass through the capillaries of the brain. These intelligent, blood-cell-sized nanobots will actually be able to go to the brain and interact with biological neurons. The basic feasibility of this has already been demonstrated in animals.

One application of sending billions of nanobots into the brain is full-immersion virtual reality. If you want to be in real reality, the nanobots sit there and do nothing, but if you want to go into virtual reality, the nanobots shut down the signals coming from my real senses, replace them with the signals I would be receiving if I were in the virtual environment, and then my brain feels as if it's in the virtual environment. And you can go there yourself — or, more interestingly you can go there with other people — and you can have everything from sexual and sensual encounters to business negotiations, in full-immersion virtual reality environments that incorporate all of the senses.

People will beam their own flow of sensory experiences and the neurological correlates of their emotions out into the Web, the way people now beam images from web cams in their living rooms and bedrooms. This will enable you to plug in and actually experience what it's like to be someone else, including their emotional reactions, a´ la the plot concept of Being John Malkovich. In virtual reality you don't have to be the same person. You can be someone else, and can project yourself as a different person.

Most importantly, we'll be able to enhance our biological intelligence with non-biological intelligence through intimate connections. This won't mean just having one thin pipe between the brain and a non-biological system, but actually having non-biological intelligence in billions of different places in the brain. I don't know about you, but there are lots of books I'd like to read and Web sites I'd like to go to, and I find my bandwidth limiting. So instead of having a mere hundred trillion connections, we'll have a hundred trillion times a million. We'll be able to enhance our cognitive pattern recognition capabilities greatly, think faster, and download knowledge.

If you follow these trends further, you get to a point where change is happening so rapidly that there appears to be a rupture in the fabric of human history. Some people have referred to this as the "Singularity." There are many different definitions of the Singularity, a term borrowed from physics, which means an actual point of infinite density and energy that's kind of a rupture in the fabric of space-time.

Here, that concept is applied by analogy to human history, where we see a point where this rate of technological progress will be so rapid that it appears to be a rupture in the fabric of human history. It's impossible in physics to see beyond a Singularity, which creates an event boundary, and some people have hypothesized that it will be impossible to characterize human life after the Singularity. My question is what human life will be like after the Singularity, which I predict will occur somewhere right before the middle of the 21st century.

A lot of the concepts we have of the nature of human life — such as longevity — suggest a limited capability as biological, thinking entities. All of these concepts are going to undergo significant change as we basically merge with our technology. It's taken me a while to get my own mental arms around these issues. In the book I wrote in the 1980s, The Age of Intelligent Machines, I ended with the spectre of machines matching human intelligence somewhere between 2020 and 2050, and I basically have not changed my view on that time frame, although I left behind my view that this is a final spectre. In the book I wrote ten years later, The Age of Spiritual Machines, I began to consider what life would be like past the point where machines could compete with us. Now I'm trying to consider what that will mean for human society.

Previous | Page 1 2 3 4 5 6 Next