Of Johns Hopkins’s amazing progress, MIT Technology Review reported in 2007, “They have demonstrated for the first time that neural activity recorded from a monkey’s brain can control fingers on a robotic hand, making it play several notes on a piano.” But this was not entirely accurate, according to Jonathan Kuniholm, a former engineer officer with the First Battalion, Twenty-third Marines. Kuniholm lost his right arm to an IED buried along the Euphrates River in Haditha, Iraq. The homemade bomb was disguised as a discarded olive oil can. After recuperating, Kuniholm signed on with DARPA and Revolutionizing Prosthetics. “The Intrinsic hand was physically capable of all the individual movements necessary to play the piano,” Kuniholm wrote in IEEE Spectrum, the trade magazine for the world’s largest professional association for the advancement of technology, “but it could not be controlled by a person in real time. There was no muscle twitch or electrical signal being decoded by signal-processing algorithms in real time. The hand was preprogrammed, like a player piano.” In some regards, Revolutionizing Prosthetics did more for DARPA’s image than it did for warfighters who had lost limbs in war.
Major news organizations wrote stories about the DEKA arm, hailing it as revolutionary, spectacular, and astounding. In 2009, Dean Kamen, DEKA’s founder, recalled on 60 Minutes what it was like when DARPA officials came to him proposing to build a robotic arm. “They said, ‘We want these kids to have something put back on them that will essentially allow one of these kids to pick up a raisin or a grape off the table, know the difference without looking at it.’” Kamen welcomed the challenge, and he and his team of forty engineers spent a year working on the problem; DARPA spent $100 million.
But when the cameras go off, the arms usually go back to the DARPA laboratories, where they generally sit on shelves. “Most of us strap back on our Captain Hook arms,” said one participant, who lost an arm in Iraq and who has appeared on national television modeling the DEKA arm but asked not to be identified by name. This individual has become frustrated with DARPA, whose motives he sees as something other than getting better prosthetics to war veterans, though he does not claim to know what DARPA’s ulterior motives might be. The DEKA arm, which costs up to $650,000 to engineer, has yet to find a partner to mass-produce its system. In November 2014 the FDA approved marketing the device, which reportedly can respond to multiple simultaneous commands from a wearer’s brain. In a press statement, DARPA said it was happy “to repay some of the debt we owe to our Service members,” but acknowledges there is no timeline on when the DEKA arm will become available to amputees. America’s wounded warriors continue to wear what amputees have worn since World War I, the so-called Captain Hook arm, which is officially called the Dorrance hook, invented by D. W. Dorrance in 1912.
It is likely that DARPA’s primary goal in advancing prosthetics is to give robots, not men, better arms and hands. Robotics expert Noel Sharkey, who serves as a United Nations advisor and chairman of the International Committee for Robot Arms Control, explains: “You hear DARPA talk about a robot they are designing, being able to turn a valve inside a Fukushima-type power plant. Yes, that is an example of robots keeping humans safe. But that robotic hand will also soon be able to turn a valve onboard, say, a ship.” A ship that a robot has been sent to take over in a military operation.
The technologies DARPA is pursuing in its brain and prosthetics programs have dual use in DARPA’s efforts to engineer hunter-killer robots. Coupled with the quest for artificial intelligence, all this might explain why DARPA is so focused on looking inside people’s brains.
High on the top of a forested plateau in the Jemez Mountains, the Los Alamos National Laboratory is a storied place with a rich and complex history of nuclear weapons research. The Los Alamos National Laboratory is also one of the largest producers of defense science in the nation, with a mission statement that reads, “Delivering science and technology to protect our nation and promote world stability.” Although the list of DARPA contracts here at Los Alamos is not public knowledge, it is voluminous. Most of the contracts are classified. These are not the programs that DARPA’s public affairs officers are quick to promote in the press. The classified programs are not like the ones people read about in mainstream magazines and newspapers, about bullets that bend, prosthetics that can pick up a grape, cars that can drive themselves, technology you can swallow, and robots that can fall down and get back up again. Here, in the classified laboratories at Los Alamos National Laboratory, and in other classified national laboratories and research facilities like this one, is where some of DARPA’s highest-risk, highest-payoff programs evolve. The consequential weapons systems of the future are born black, as in classified, and, like the hydrogen bomb, McNamara’s electronic fence, Assault Breaker, and stealth technology, are unveiled to the public only after they have created a revolution in military affairs.
Within the thirty-six-square-mile Los Alamos campus, there are 1,280 buildings, eleven of which are nuclear facilities. Even the cooks who work in some of the kitchens have top secret Q clearances. There are sixty-three miles of gas lines inside the laboratory campus, thirty-four miles of electrical lines, and a power plant. There are roughly ten thousand employees and contract workers at the lab, and according to the historian at the Los Alamos Historical Society, roughly half of them have Ph.D.s. One scientist who has a DARPA contract and is at liberty to discuss some of his work on the artificial brain is Dr. Garrett T. Kenyon.
Outside Dr. Kenyon’s office at Los Alamos there is an armored truck with a machine gun mounted on top. It is parked in the red zone, by the front entrance. Inside the building, Dr. Kenyon and his team work on artificial intelligence, man’s quest to create a sentient machine. Dr. Kenyon is part of the synthetic cognition group at Los Alamos National Laboratory. He and his team are simulating the primate visual system, using a supercomputer to power the operation. Specifically, the team is trying to create a precise computer model of the human eye, including all of its neural networks, to understand the relationship between visual cognition and the brain. This is not necessarily an impossible task, but it does require one of the fastest computers in the world to model such a complex neural network as that of the human eye. Neuroscientists currently believe that there are 100 billion neurons inside a human brain and that every sensory message the brain receives involves an exponential number of neural connections between these networks.
To do their work, Dr. Kenyon and his team use a part of the IBM Roadrunner supercomputer, or what is left of it. When Roadrunner was built in 2008 it was the fastest computer in the world, able to perform 1 million billion calculations per second, setting the world’s record for petaflops per second data-processing speeds. That is a far cry from the World War II–era ENIAC computer at the University of Pennsylvania’s Moore School, which completed five thousand operations per second. But science builds. Visions become reality. Thus the ENIAC inspired John von Neumann to build MANIAC, which inspired Daniel Slotnick to build the ILLIAC IV, which led to the IBM Roadrunner. In 2014, the world’s fastest supercomputer, located at China’s National University of Defense Technology and called Tianhe-2, could reportedly perform some 30 quadrillion calculations per second, or 33.86 petaflops.
As for the IBM Roadrunner supercomputer, between its unveiling in 2008 and my visit in 2014, it has become obsolete. The machine cost $100 million to build but has since become too power-inefficient to continue to run. The machine cannot be recycled, though, because it holds many of the nation’s nuclear secrets. Computers never entirely lose the information they record. Because of this, and since the Los Alamos National Laboratory requires a bigger, faster, more efficient computer, Roadrunner is being destroyed. Some of what is left of it is being used by Dr. Kenyon’s team in their quest for artificial intelligence. The banks of computers they use fit into a room about the size of a basketball court.
Dr. Kenyon takes me to look at the supercomputer. It is located inside the brick and glass building that houses his laboratory, beyond t
he armored truck, down a long corridor and behind a single locked door. Dr. Kenyon and I peer in through a small window at the Roadrunner supercomputer. The lights are low. The banks of processors are alight with tiny red and white blinking lights. There are racks of machines in rows. There are bundles of cables on the floor. Kenyon points inside. “It’s a giant abacus,” he says. “The real power is in the human brain.” Kenyon taps his forehead. “So small, so infinite.”
We walk through another part of the building. While we wait for an elevator, Dr. Kenyon unfolds a dinner-size napkin and holds it up in the air in front of his forehead. “This is about the size of your brain, spread out,” he says. “The part that matters. The cerebral cortex.” The 100 billion neurons there are also known as the brain’s gray matter. “And the human brain does things beyond anyone’s comprehension. Evolution created the smartest machine in this world.”
Dr. Kenyon explains the concept behind the DARPA-funded project he is working on, in layman’s terms. “Today, my twelve-year-old daughter reprogrammed my smart phone so it has facial recognition software,” he says. “But seventy to eighty percent of the time it doesn’t recognize me.” He holds up his phone to his face. “The smart phone can’t always see it’s me. I can see it’s me. There’s the double chin, like it or not. So why can’t my phone recognize me all the time? Why can’t it perform a function that my dog can, the minute I walk in the door? For all the things the smart phone can do, it can’t do the simplest things that biological systems can. Recognize someone all the time.”
Kenyon notes that if a person’s teenaged child recognized him only 70 to 80 percent of the time, there would be something seriously wrong with the child’s brain. “Sentient beings recognize through sight,” he explains. “My phone, on the other hand, is just comparing a set of stored features with a set of features extracted from the input coming from its camera. It’s not ‘seeing’ anything. My phone is not resolving the pixels into a rich scene, with all the interrelationships implicit therein. My phone is just finding a few key points and constructing a high-dimensional feature vector that it can compare to a stored feature vector.”
At present, true recognition—as in cognition, or acquiring knowledge and understanding through thought, experience, and the senses—is done only by sentient beings. “We think that by working hard to understand how biological systems solve this problem, how the primate visual system recognizes things, we can understand something fundamental about how brains solve the problems they do, like recognition. Until then, computers are blind,” Kenyon says. “They can’t see.”
Which raises at least one technical problem regarding artificial intelligence and autonomous hunter-killer drones. “I think robot assassins are a very bad idea for a number of reasons,” Garrett Kenyon asserts. “Moral and political issues aside, the technical hurdles to overcome cannot be understated,” he says. “It’s misleading to think just because my smart phone can ‘identify’ me seventy percent of the time that it has thirty percent to go.” We are talking about orders of magnitude. “The chances that my daughter might not recognize me, or misidentify me from a short distance, or because I am wearing a hat,” he says, “are about one in 0.0001. And we still do not understand how neural systems work.”
Dr. Kenyon is excited by his research. He is convinced that neuroscientists of today are like alchemists of the Middle Ages trying to understand chemistry. That all the exciting discoveries lie ahead. “Think of how much chemists in the Dark Ages did not understand about chemistry compared to what we know now. We neuroscientists are trapped in a bubble of ignorance. We still don’t have a clue about what’s going on in the human brain. We have theories; we just don’t know for sure. We can’t build an electrical circuit, digital or analogue or other, that mimics the biological system. We can’t emulate the behavior. One day in the future, we think we can.”
Dr. Kenyon says that one of the most powerful facts about DARPA as an organization is that it includes theoretical scientists and engineers in its ranks. The quest for artificial intelligence, he says, is similar to getting humans to Mars. Once you have confidence you can do it, “then getting to Mars is an engineering problem,” he says. In his laboratory, metaphorically, “we just don’t know where Mars is yet.” But Dr. Kenyon and his team are determined. “I don’t think it’s that far away,” he says of artificial intelligence. “The question is, who will be the Columbus here?”
Columbus was an explorer looking for a new land. DARPA is looking for ways to use science to fight future wars.
Interviews with DARPA scientists of today give a sense that in the twenty-first century, programs that once existed in the realm of science fiction are rapidly becoming the science of the here and now. If Dr. Garrett Kenyon’s Los Alamos laboratory represents the future of the mind, the laboratory of Dr. Susan V. Bryant and Dr. David M. Gardiner at the University of California, Irvine, represents the future of the human body. Dr. Bryant and Dr. Gardiner are a husband-and-wife team of regeneration biologists. Dr. Bryant also served as the dean of the School of Biological Sciences and the vice chancellor for research at U.C. Irvine. Dr. Gardiner is a professor of developmental and cell biology and maintains the laboratory where he does research as a regenerative engineer.
This laboratory looks like many university science labs. It is filled with high-powered microscopes, dissection equipment, and graduate students wearing goggles and gloves. The work Dr. Gardiner and Dr. Bryant do here is the result of a four-year contract with DARPA and an extended five-year contract with the Army. Their work involves limb regeneration. Gardiner and Bryant believe that one day soon, humans will also be able to regenerate their own body parts.
Dr. David Gardiner, who is in his sixties, examines a set of lab trays on the countertop. Crawling around inside the trays are multi-limbed aquatic salamanders called axolotls. The creatures look both prehistoric and futuristic, with large, bug-like eyes. Some are pink; others are unpigmented, a naturally occurring mutation that makes them look transparent; you can see the bones and blood vessels inside. This species of salamander, a urodele amphibian, is able to regenerate lost body parts as an adult.
“Regeneration is really coming alive now,” Dr. Gardiner says. “Sue and I have been studying the science for years. DARPA was the first time anyone ever asked us to regenerate anything. They did this with the mouse digit,” he says, referring to the tip of a mouse finger, which they and another team of scientists had been able to get to grow back, thereby setting a scientific milestone. “DARPA said, ‘Great. Can you scale it up?’ As in pigs. As in humans. They asked, ‘Is this possible?’ We said yes. They asked, ‘Do you know how to do it?’ We said no. They said, ‘Well, then, we’ll fund you.’” Gardiner believes that therein lies the genius of DARPA. “DARPA funds questions,” he says.
Dr. Gardiner searches through the trays of salamanders and locates the one he is looking for. This axolotl has an extra limb coming out the right side of its body. A second right front limb. “If we look at this extra limb on the salamander, we understand we [humans] have all the info to make an arm.”
To explain the concept of limb regeneration, Dr. Gardiner first provides a brief summary of mutagenesis, the process by which an organism’s genetic information is changed, resulting in a mutation. “Mutations occur in nature, as the result of exposure to a mutagen,” he says. “Natural mutations can be beneficial or harmful to an organism, and this drives evolution. Mutations can also be performed as experiments, in laboratories. DNA can be modified artificially, by chemical and biological agents, resulting in mutations.” One consequential example of harmful mutagenesis that we discuss occurred as a result of ARPA’s Project Agile defoliation campaign. People who were exposed to Agent Orange during the Vietnam War suffer a higher rate of children born with mutations. This includes Vietnamese people who were sprayed with the herbicides and also a vigorously debated number of American servicemen who were involved in the spraying.
“Mutations tell us about signals
,” Gardiner explains. “Cells talk to each other using signals. Every cell has an identity. All cells have information. There are no dumb cells. Cells talk to each other to stimulate growth. They talk to each other to make new patterns.” Pointing to the see-through axolotl with the extra limb, Gardiner says, “People look at this salamander and say, ‘Salamanders are special. We [humans] will never regenerate like a salamander.’” Dr. Gardiner and Dr. Bryant do not agree. “We say, ‘Oh, really? How do you know?’ The most compelling evidence is you have an arm.”
There is no regeneration gene, says Gardiner. It happens at a cellular level. “People regenerate. Look how we started ab initio. As a single cell. Once upon a time, each one of us was a one-cell embryo that divided. Every human being on this planet regenerated his or her own cells, in the womb.”
Dr. Bryant uses differentiation to simplify things. “The difference between salamanders and humans,” she says, “is that when salamanders’ limbs are amputated, they grow new ones. When humans’ limbs are amputated, they produce scar tissue. We humans respond to injury by making scar tissue. Why?” she asks.
“At the heart of limb regeneration is evolution,” Dr. Gardiner adds. What his wife is pointing out, he says, is that “at the heart of genetics is diversity.”
“Some people make mega-scars,” says Dr. Bryant. “The scars can be bigger than the wound. If you cut the scar tissue off, it grows back. There is the same evidence at the other end of the scarring spectrum. Some people produce scars that can go away.”
Dr. Gardiner suggests looking at cancer research as an analogy. “Cancer equals our bodies interacting with the environment,” he says. “Cancer shows us we have remarkable regenerative ability. The pathways that drive cancer are the same pathways that cause regeneration. In the early days, no one had any idea about cancer. There was one cancer. Then along came the idea of ‘cancer-causing’ carcinogens. Well, we have found salamanders are very resistant to cancer. Inject a carcinogen into a salamander and it regulates the growth and turns it into an extra limb.”