Chapter 4: Thinking
78 - Evolutionary Robotics (Part Three)
Asked if computers will succeed us, MIT’s turns the question on its head. “You can’t know, because it’s so hypothetical,” he says. “But I think to a certain extent we’re going to have to face issues like this a lot earlier than we think. But it’s going to be the cyborg, the melding of silicon and living things. That’s where these issues are going to occur first. It’s not going to be purely intelligent robot versus current existing life form…It’s going to be humans mixed with silicon, and the basic intelligence is still going to be human intelligence.
“So mind you, once we get through that, if we get through that, or how we get through that, who knows what it’s going to be later on? Because just that is going to be such a shattering debate for what humankind is. It’s going to transform the whole landscape so much, and it’s coming.
“The intelligent robot that is as good as a human, or supersedes a human, I don’t think I’m going to see that. I’d love to, I mean that’s what my life’s work is dedicated to. But realistically I don’t think I’m going to see that…Not that we won’t have really intelligent machines, but they aren’t going to be introduced into this world. They are going to be introduced into a transformed world that we can’t even begin to imagine.”
Toward that end a UC San Diego team, led by physicist Henry Abarbanel, has connected lobster neurons to a circuit they made using $7.50 worth of parts bought at RadioShack. According to Abarbanel, the living cells and electronic circuit worked smoothly as a single unit. That work is supplemented by advances in “neurocomputing” by Georgia Tech’s neurocomputing and others. The University of Florida’s Thomas DeMarse has taught a dish filled with 25,000 living rat neurons laid on a grid of sixty electrodes to fly an F-22 fighter jet simulator. And at Arizona State University in Tempe and the Neurosciences Institute near San Diego, researchers have implanted electrodes in the brains of monkeys, who can now move a robotic hand and a cursor on a computer screen with only their thoughts. Companies with names like Advanced Bionics, Cyberkinetics, and Neural Signals (led by pioneering neuroscientist Philip Kennedy) are working with human paralytics, aiming to give them similar abilities.
With all this activity, can Brooks’ cyborg mind meld be far off? A group at Cornell has used DARPA funding to develop a bio-silicon Hybrid Information Appliance that instead of relying on batteries can be fed with sugar in the field, and DARPA has awarded millions to the Brain-Machine Interface Program.
As artificial intelligence continues to evolve, it feeds into yet another new field: “complexity.” If AI aims to mimic the dynamics of individual minds, the science of complexity looks to a broader dynamic—that of large numbers of individuals interacting in complex systems. How do order and chaos work together in sustaining life? Why do higher levels of organization keep emerging in nature? At places like the Santa Fe Institute, out in the rugged landscape of the American Southwest, scientists are coming up with provocative new clues.
Email this to a friend | 
Add to del.icio.us | 
Add to digg |
Linkology by James (all links are unofficial) | Rights Info
