Integral World: Exploring Theories of Everything
An independent forum for a critical discussion of the integral philosophy of Ken Wilber
Andrea Diem-LaneAndrea Diem-Lane is a tenured Professor of Philosophy at Mt. San Antonio College, where she has been teaching since 1991. Professor Diem has published several scholarly books and articles, including The Gnostic Mystery and When Gods Decay. She is married to Dr. David Lane, with whom she has two children, Shaun-Michael and Kelly-Joseph.

The Artificial Brain
of the Future

Looking at Ray Kurzweil's ‘How to Create a Mind’

Andrea Diem-Lane

How to Create a Mind

Ray Kurzweil in How to Create a Mind offers the reader a glimpse of the future, and not too distance one at that. By 2029 Kurzweil predicts that we will have computers with consciousness, or at least “convincingly” so. And arguably by the 2030s non-biological minds, with an emotional, moral and intellectual makeup, may be commonplace. Certainly, we will have to take the “leap of faith” that conscious computers have qualia, in the same we take the leap of faith that another human does (at this stage there is no way for me to truly know the mind of another).  Not only will these machines be able to pass the Turing Test, perhaps having had to dumb themselves down a bit to appear more human like, but they will be conscious beings we can laugh with, cry with, and even get mad out. As such we will see them not as wires and chips, but as entities we can develop empathy with and even care for.

The human mind and the computer will be indistinguishable, Kurzweil contends, except with the caveat that the computer version will be so much faster (10 million times faster than the brain's electrochemistry) and much more capable of processing data. We know that the human brain has limitations, being able he says to hold only 4 simultaneous lists at the same time, but the computer version will be nearly limitless. It can access billions of data, including vast literature and material from disconnected fields of study, to further its advanced understanding. And it will be able to share this knowledge with other computers immediately. Furthermore, a critical thinking and moral component can be built into it, and this can contribute to building a better world.

In his chapter on “Transcendent Abilities” morality, love, and creativity are discussed. Kurzweil analyzes what is going on inside the human brain that can account for such abilities, with the intention that we can utilize this knowledge and apply it to creating emotions and “transcendent abilities” in computers. For instance, when it comes to morality, he points to spindle neurons, the largest neurons we have in the brain, to understand moral judgement. Humans possess 80,000 spindle neurons (45,000 in the right hemisphere and 35,000 in the left hemisphere) and they start to develop around 4 months old. Gorillas, bonobos, and chimpanzees also have these cells, but at a much lower level. This shows the rudiments of morality in our closest ancestors. Love is another feature he examines from a biochemical perspective. When we fall in love dopamine is released, norepinephrine increases and serotonin decreases, similar to fight or flight reactions. Oxytocin and vasopressin also play a pivotal role (see the research on praire voles vs. montane voles to really appreciate the role of these neurotransmitters on love and mate bonding). Kurzweil suggests that once we understand the biochemistry of the human brain perhaps computers may be programmed in a similar way.

There is a wealth of evidence Kurzweil points to in this text to demonstrate that conscious computers are on their way. At Harvard one researcher has simulated a round worm brain and uploaded it to the virtual world. Another researcher connected to the Blue Brain Project has reverse engineered an entire section of a rat's brain and predicts that by 2023 the human brain will be fully completed. In fact, a team connected to IBM has already finished part of the human visual cortex. While we may be at the early stages of this endeavor, advancements will be made rapidly. Kurzweil coins the term LOAR, or law of accelerated returns, to highlight the exponential growth in biotechnology and in technology in general. Couple this with PRTM, or pattern recognition theory of the mind, (this refers to understanding the algorithm of the neocortex with its 300 million pattern processors), and we can expect that within the near future we will witness conscious machines. Understanding the hierarchical thinking of the neocortex and the redundancy built into the brain will enable us to carry this over to the digital world and huge advancement may be made.

Not everyone agrees with Kurzweil's predictions and in his text he discusses his critics, including Roger Penrose. According to Penrose, the human brain possesses “transcendent capabilities” due to neural quantum effects of the microtubles and so is unique to the human and it cannot be replicated in a computer. Kurzweil clearly disagrees with this, asserting that the distinction between a human and a computer “will blur until the differences disappear.”

John Searle, famous for his Chinese Room argument, is also mentioned. While one might think of Searle as opposing Kurzweil, since he is known for rejecting strong AI and a reductionist view, the writer interestingly states that they are not necessarily on the opposite ends. He quotes Searle: “The first step is to figure out how the brain does it and then build and artificial machine that has an equally effective mechanism for causing consciousness.” Yet, having read Searle's Mystery of Consciousness, he painstakingly objects to Daniel Dennett's strong AI position and I am not so sure Kurzweil can really claim him as a supporter. According to the Chinese Room argument the one behind the door who manipulates symbols and slips under the door correct answers to questions in Chinese may appear to understand Chinese but does not. This is how a computer works, contends Searle. It can simulate consciousness but not have consciousness. In other words, the computer can play with syntax and appear to be conscious, but it is not since it truly lacks semantics. Kurzweil counters this with a compelling response. He says we can apply this reasoning to the human brain as well, as the neurons are simply following an algorithm much like a computer is following a program. Kurzweil is not arguing here that humans lack understanding but that the Chinese Room argument is not a convincing argument.

In this reading, Kurzweil clarifies that he is not disparaging humans in any way, reducing us to less than a machine. Rather, this futurist claims he is elevating our understanding, not just of future machines but of our very selves. Instead of predicting a gloom and doom scenario, he seems to paint a bright prospect for us. In the Epilogue of the text, he brings up several examples that illustrates an upward progression. For instance, life expectancy has climbed 4 fold in the last 1000 years and a free democracy is more of the norm today than ever before. Add to the advancements an intelligence explosion, or Singularity moment Kurzweil predicts by 2045 (the Singularity term was first coined by John von Neumann in the 1950s), and a golden age seems at await us.

In addition to creating a computer that can think and feel, Kurzweil predicts that we will be able to repair our own brains digitally. When a part in our human brain is not working we can implant a computer component to remedy the situation. As Alzheimer's destroys the hippocampus, a digital implant to this region of the neocortex can ensure our memories stay intact. If our thalamus, an essential component to process sensory information into our neocortex, is damaged we can replace it. Thus, we can repair ourselves and preserve our consciousness. Moreover, besides repairing damaged neural components, new biotechnology will allow us to significantly enhance our own intelligence as we add digital parts to our brains. Brain augmentation, he suggests, will be prevalent.

Ray Kurzweil
Ray Kurzweil

A very captivating idea Kurzweil proposes in this reading is the You 2 argument. Through a thought experiment, he asks us to entertain the idea of creating another you. Perhaps while you are asleep a research team utilizing non-invasive technology (maybe through “blood cell-sized scanning machines travelling the capillaries of your brain”) is able to detail your brain and load this “mind-file” in a non-biological being. The cloned mind is what he calls You 2. Many of us may shudder at the thought of another you out there. Or instead of making another you, perhaps each part of your brain is slowly but surely replaced with a more enhanced digital part so that eventually all of you is replaced and you are now the You 2 itself. Again, many may feel another nervous shiver down the spine. Yet, Kurzweil points out that there already is a You 2 being creating constantly as our cells are continuously being replaced. For instance, cells in the stomach turn over in weeks, blood cells lasts days to months, and even neuron's organelles lasts a month. Thus, in a matter a short time another You has been created. And somehow this news of the shedding and replacing of our cells does not freak us out. So why, this futurist ponders, do we resist so much a digital You 2.

Predicting the future is the reason we evolved our brains, argues Kurzweil. In my text Darwin's DNA I present a thesis quite similar to this: consciousness resulted from the virtual simulation of our ancestors. Those who were able to imagine, or virtually simulate, the lion behind the bush when the bush rustled had a better chance at survival than those who did not. Even if they were only right 1 time out of 100, it was this one time that meant their genes stayed in the mix. So in way our early ancestor were making predictions of a potential future event that might occur and such predictions could be life-saving. This virtual simulation trait, perhaps the foundation of human consciousness, was naturally selected for and we are the descendants who inherited it.

The general claim of evolutionary biologists is that we are not evolving with a goal or a direction. Steven Pinker says there is no striving for intelligence. Yet, intriguingly, Kurzweil begs to differ here when it comes to the evolution of the human neocortex. In the section on the evolution of our brains, Kurzweil challenges Pinker's position arguing that we are, in fact, moving in the direction of greater intelligence. Much like when we release a cap off of a jar filled with gas, the molecules move in all directions and some of the molecules move up and out. Similarly, not all evolution is in the direction of greater intelligence, but once hierarchical learning did evolve it was such an important survival advantage it continually was selected for. The end result is a Homo sapien with a large neocortex.

Knowing that envisioning the future is in our genes, it makes sense why Kurzweil is so fascinated with writing about it, and why I am equally engrossed in reading his predictions. Repairing, augmenting, or replacing our neocortex or creating new minds in conscious machines may be right around the corner. “Waking up the universe” with our intelligence and that of the non-biological form is a glorious future, he concludes, and it is “our destiny.”

Comment Form is loading comments...