The Future of the Mind (40 page)

Now assume that such a drone suffers a breakdown, such that its facial recognition software malfunctions. Then it becomes a rogue robot, with permission to kill anyone in sight. Worse, imagine a fleet of such robots controlled by a central command. If a single transistor were to blow out in this central computer and malfunction, then the entire fleet might go on a killing spree.

A more subtle problem is when robots perform perfectly well, without any malfunctions, yet there is a tiny but fatal flaw in their programming and
goals. For a robot, self-preservation is one important goal. But so is being helpful to humans. The real problem arises when these goals contradict each other.

In the movie
I, Robot
, the computer system decides that humans are self-destructive, with their never-ending wars and atrocities, and that the only way to protect the human race is to take over and create a benevolent dictatorship of the machine. The contradiction here is not between two goals, but within a single goal that is not realistic. These murderous robots do not malfunction—they logically conclude that the only way to preserve humanity is to take control of society.

One solution to this problem is to create a hierarchy of goals. For example, the desire to help humans must outrank self-preservation. This theme was explored in the movie
2001
. The computer system HAL 9000 was a sentient computer capable of conversing easily with humans. But the orders given to HAL 9000 were self-contradictory and could not be logically carried out. By attempting to execute an impossible goal, it fell off the mesa; it went crazy, and the only solution to obeying contradictory commands from imperfect humans was to eliminate the humans.

The best solution might be to create a new law of robotics, which would state that robots cannot do harm to the human race, even if there are contradictions within their previous directives. They must be programmed to ignore lower-level contradictions within their orders and always preserve the supreme law. But this might still be an imperfect system at best. (For example, if the robots’ central goal is to protect humanity to the exclusion of all other goals, then it all depends on how the robots define the word “protect.” Their mechanical definition of this word may differ from ours.)

Instead of reacting with terror, some scientists, such as Dr. Douglas Hofstadter, a cognitive scientist at Indiana University, do not fear this possibility. When I interviewed him, he told me that robots are our children, so why shouldn’t we love them like our own? His attitude, he told me, is that we love our children, even though we know that they will take over.

When I interviewed Dr. Hans Moravec, former director of the AI Laboratory at Carnegie Mellon University, he agreed with Dr. Hofstadter. In his book
Robot
, he writes, “
Unleashed from the plodding pace of biological evolution, the children of our minds will be free to grow to confront immense and fundamental challenges in the larger universe.… We humans will benefit
for a time from their labors, but … like natural children, they will seek their own fortunes, while we, their aged parents, silently fade away.”

Others, on the contrary, think that this is a horrible solution. Perhaps the problem can be solved if we make changes in our goals and priorities now, before it is too late. Since these robots are our children, we should “teach” them to be benevolent.

FRIENDLY AI

Robots are mechanical creatures that we make in the laboratory, so whether we have killer robots or friendly robots depends on the direction of AI research. Much of the funding comes from the military, which is specifically mandated to win wars, so killer robots are a definite possibility.

However, since 30 percent of all commercial robots are manufactured in Japan, there is another possibility: robots will be designed to become helpful playmates and workers from the very beginning. This goal is feasible if the consumer sector dominates robotics research. The philosophy of “friendly AI” is that inventors should create robots that, from the very first steps, are programmed to be beneficial to humans.

Culturally, the Japanese approach to robots is different from the West’s. While kids in the West might feel terror watching rampaging Terminatortype robots, kids in Japan are steeped in the Shinto religion, which believes spirits live in all things, even mechanical robots. Instead of being uncomfortable at the sight of robots, Japanese children squeal with delight upon encountering them. It’s no wonder, therefore, that these robots in Japan are proliferating in the marketplace and in homes. They greet you at department stores and educate you on TV. There is even a serious play in Japan featuring a robot. (Japan has another reason for embracing robots. These are the future robot nurses for an aging country. Twenty-one percent of the population is over sixty-five, and Japan is aging faster than any other nation. In some sense, Japan is a train wreck in slow motion. Three demographic factors are at work. First, Japanese women have the longest life expectancy of any ethnic group in the world. Second, Japan has one of the world’s lowest birthrates. Third, it has a strict immigration policy, with over 99 percent of the population being pure Japanese. Without young immigrants to take care of the elderly, Japan may rely on robot nurses. This problem is not restricted
to Japan; Europe is next. Italy, Germany, Switzerland, and other European nations face similar demographic pressures. The populations of Japan and Europe could experience severe shrinkage by mid-century. The United States is not far behind. The birthrate of native-born U.S. citizens has also fallen dramatically in the last few decades, but immigration will keep the United States expanding into this century. In other words, it could be a trilliondollar gamble to see if robots can save us from these three demographic nightmares.)

Japan leads the world in creating robots that can enter our personal lives. The Japanese have built robots that can cook (one can make a bowl of noodles in a minute and forty seconds). When you go to a restaurant, you can place your order on a tablet computer and the robot cook springs into action. It consists of two large, mechanical arms, which grab the bowls, spoons, and knives and prepare the food for you. Some robotic cooks even resemble human ones.

There are also musical robots for entertainment. One such robot actually has accordion-like “lungs” by which it can generate music by pumping air through an instrument. There are also robot maids. If you carefully prepare your laundry, it can fold it in front of you. There is even a robot that can talk because it has artificial lungs, lips, tongue, and nasal cavity. The Sony Corporation, for example, built the AIBO robot, which resembles a dog and can register a number of emotions if you pet it. Some futurists predict that the robotics industry may one day become as large as the automobile industry is today.

The point here is that robots are not necessarily programmed to destroy and dominate. The future of AI is up to us.

But some critics of friendly AI claim that robots may take over not because they are aggressive, but because we are sloppy in creating them. In other words, if the robots take over, it will be because we programmed them to have conflicting goals.

“I AM A MACHINE”

When I interviewed Dr. Rodney Brooks, former director of the MIT Artificial Intelligence Lab and cofounder of iRobot, I asked him if he thought machines would one day take over. He told me that we just have to accept
that we are machines ourselves. This means that one day, we will be able to build machines that are just as alive as we are. But, he cautioned, we will have to give up the concept of our “specialness.”

This evolution in human perspective started with Nicolaus Copernicus when he realized that the Earth was not the center of the universe, but rather goes around the sun. It continued with Darwin, who showed that we were similar to the animals in our evolution. And it will continue into the future, he told me, when we realize that we are machines, except that we are made of wetware and not hardware.

It’s going to represent a major change in our world outlook to accept that we, too, are machines, he believes. He writes, “
We don’t like to give up our specialness, so you know, having the idea that robots could really have emotions, or that robots could be living creatures—I think is going to be hard for us to accept. But we’re going to come to accept it over the next fifty years.”

But on the question of whether the robots will eventually take over, he says that this will probably not happen, for a variety of reasons. First, no one is going to accidentally build a robot that wants to rule the world. He says that creating a robot that can suddenly take over is like someone accidentally building a 747 jetliner. Plus, there will be plenty of time to stop this from happening. Before someone builds a “super-bad robot,” someone has to build a “mildly bad robot,” and before that a “not-so-bad robot.”

His philosophy is summed up when he says, “The robots are coming, but we don’t have too much to worry about. It’s going to be a lot of fun.” To him, the robot revolution is a certainty, and he foresees the day when robots will surpass human intelligence. The only question is when. But there is nothing to fear, since we will have created them. We have the choice to create them to help, and not hinder, us.

MERGE WITH THEM?

If you ask Dr. Brooks how we can coexist with these super-smart robots, his reply is straightforward: we will merge with them. With advances in robotics and neuroprosthetics, it becomes possible to incorporate AI into our own bodies.

Dr. Brooks notes that the process, in some sense, has already begun. Today, about twenty thousand people have had cochlear implants, which
have given them the gift of hearing. Sounds are picked up by a tiny receiver, which converts sound waves to electrical signals, which are then sent directly to the auditory nerves of the ear.

Similarly, at the University of Southern California and elsewhere, it is possible to take a patient who is blind and implant an artificial retina. One method places a mini video camera in eyeglasses, which converts an image into digital signals. These are sent wirelessly to a chip placed in the person’s retina. The chip activates the retina’s nerves, which then send messages down the optic nerve to the occipital lobe of the brain. In this way, a person who is totally blind can see a rough image of familiar objects. Another design has a light-sensitive chip placed on the retina itself, which then sends signals directly to the optic nerve. This design does not need an external camera.

This also means that we can go even further and enhance ordinary senses and abilities. With cochlear implants, it will be possible to hear high frequencies that we have never heard before. Already with infrared glasses, one can see the specific type of light that emanates from hot objects in the dark and that is normally invisible to the human eye. With artificial retinas, it may be possible to enhance our ability to see ultraviolet or infrared light. (Bees, for example, can see UV light because they have to lock onto the sun in order to navigate to a flower bed.)

Some scientists even dream of the day when exoskeletons will have superpowers like those found in comic books, with super strength, super senses, and super abilities. We’d become a cyborg like Iron Man, a normal human with superhuman abilities and powers. This means that we might not have to worry about super-intelligent robots taking over. We’d simply merge with them.

This, of course, is for the distant future. But some scientists, frustrated that robots are not leaving the factory and entering our lives, point out that Mother Nature has already created the human mind, so why not copy it? Their strategy is to take the brain apart, neuron by neuron, and then reassemble it.

But reverse engineering entails more than just creating a vast blueprint to create a living brain. If the brain can be duplicated down to the last neuron, perhaps we can upload our consciousness into a computer. We’d have the ability to leave our mortal bodies behind. This is beyond mind over matter. This is mind without matter.

I’m as fond of my body as anyone, but if I can be 200 with a body of silicon, I’ll take it.

—DANIEL HILL, COFOUNDER OF THINKING MACHINES CORP
.

11
REVERSE ENGINEERING THE BRAIN

In January 2013, two bombshells were dropped that could alter the medical and scientific landscape forever. Overnight, reverse engineering the brain, once considered to be too complex to solve, suddenly became a focal point of scientific rivalry and pride between the greatest economic powers on Earth.

First, in his State of the Union address, President Barack Obama stunned the scientific community by announcing that federal research funds, perhaps to the tune of $3 billion, might be allocated to the Brain Research Through Advancing Innovative Neurotechnologies (or BRAIN) Initiative. Like the Human Genome Project, which opened the floodgates for genetic research, BRAIN will pry open the secrets of the brain at the neural level by mapping its electrical pathways. Once the brain is mapped, a host of intractable diseases like Alzheimer’s, Parkinson’s, schizophrenia, dementia, and bipolar disorder might be understood and possibly cured. To jump-start BRAIN, $100 million might be allocated in 2014 toward the project.

Almost simultaneously, the European Commission announced that the Human Brain Project would be awarded 1.19 billion euros (about $1.6 billion) to create a computer simulation of the human brain. Using the power
of the biggest supercomputers on the planet, the Human Brain Project will create a copy of the human brain made of transistors and steel.

Proponents of both projects stressed the enormous benefits of these endeavors. President Obama was quick to point out that not only would BRAIN alleviate the suffering of millions of people, it will also generate new revenue streams. For every dollar spent on the Human Genome Project, he claimed, about $140 of economic activity was generated. Entire industries, in fact, sprouted with the completion of the Human Genome Project. For the taxpayer, BRAIN, like the Human Genome Project, will be a win-win situation.