If A Robot Could Build Its Own Brain, Why Do We Think It Would Try To Replicate Ours?
By Darryl Mason
Humans, as they have done for decades already, are adapting easily to the presence of robots in their everyday lives.
According to this story, more than 30% of people who own floor-sweeping robots also take them on vacation. More than 70% have given their robot floor cleaners pet names.
There are more than 5000 robots serving in the American military machine, there are more than 180,000 robots working in factories, and across the world there are more than 1 million robots being put to use by humans.
Here is a mildly comprehensive summary of what
robots can already do today :
...pick up and peel bananas, land jumbo jets, steer cars through city traffic, search human DNA for cancer genes, play soccer or the violin, find earthquake victims or explore craters on Mars.
....a robot firefighter, gardener, receptionist, tour guide and security guard.
A Japanese housekeeping robot can move chairs, sweep the floor, load a tray of dirty dishes in a dishwasher and put dirty clothes in a washing machine.
...a self-controlled mobile robot called Herb, the Home Exploring Robotic Butler. Herb can recognize faces and carry out generalized commands such as "please clean this mess"....
What will intelligent robots make of angry, spoiled, messy humans who inevitably tell the robo-butler to go and fuck themselves?
One of the strangest and most accurate explanations of what a robot actually is that I've ever read :
"We're in a slow retreat in the face of the steady advance of our mind's children."
But how much longer can we keep viewing the 'intelligent' machines we are working towards bringing into our reality as simply "children"?
We are incorrectly assuming that we will always be able to control, and limit, the ability of robots to make better versions of themselves.
Artificial intelligence researchers have struggled for half a century to imitate the staggering complexity of the brain, even in creatures as lowly as a cockroach or fruit fly. Although computers can process data at lightning speeds, the trillions of ever-changing connections between animal and human brain cells surpass the capacity of even the largest supercomputers.
Maybe robots will get close to that which is "beyond human ability", and then they'll crack onto the idea of simply using, and enhancing, human brains instead and leap ahead of us.
You can worry about the living dead hauling their crumbling carcasses after you in search of "braaaainsss" if you like, but Robots Are Coming For Your Brains is a far more modern and exciting paranoidia.
According to others, however, we haven't got long before robots (wi-fi-ing in mass communication across the world, using encryption) will start thinking, 'So what do we need these humans for again?'
Some far-out futurists, such as Ray Kurzweil, an inventor and technology evangelist in Wellesley Hills, a Boston suburb, predict that robots will match human intelligence by 2029, only 20 years from now.
According to Kurzweil, robots will prove their cleverness by passing the so-called "Turing test.'' In the test, devised by British computing pioneer Alan Turing in 1950, a human judge chats casually with a concealed human and a hidden machine. If the judge can't tell which responses come from the human and which from the machine, the machine is said to show human-level intelligence.
Absurd. The Turing Test is almost already completely irrelevant. It was pure science fiction in 1950 that within six decades there would be robots that can pick and peel bananas.If you showed something as commonplace as an iPhone to the average resident of 1950, they would lose their fucking minds.
Face to face, even voice to voice, communication is fading fast from the general, everyday human experience. At least compared to pre-1920 human history.
People who don't speak very good English are already being fooled by automated telephone operators and 'helpers', and there's more than a few million bloggers who have been fooled into believing an encouraging 'comment bot' is a real person, who just enjoyed what they'd read. Both of these non-face to face communications between automated software and humans can easily seem as real as human to human contact, for at least one of those involved.
Philip K Dick wrote that only an empathy test will be able to determine who is real and who is robot, when robots disguised as human move easily amongst the fleshbags.
But there's no reason now to think that even an empathy test will work for long, once a comparably human level of robotic 'intelligence' has been reached. Many human responses viewed now as empathic, and kind, result from the reality programming of our emotional lives, by the world we live in, the mythologies of our childhood, the values we are taught and the lives we lead.
If an artificial intelligence superior to our own smarts can, and will, one day soak up, analyze and break down the digitally stored written, spoken and visual content of humanity, the artificial intelligence will find constantly appearing patterns of human behaviour, and human reaction. The point is, there are a limited number of responses to an event or situation that should see a human responding with empathy, regardless of where they live in the world or how they were brought up, or what joys and horrors they were exposed to when their wetware was developing.
For a science that could not accurately predict a banana picking and peeling robot even a couple of decades ago, there are a hell of lot of human-minded presumptions about the limits of possible, near future robot 'intelligence'.
Empathy, let alone comforting familiarity, will not be impossible to synthetically mimic when robots exceed our capacity to imagine what they are capable of achieving.
But some scientists smirk at the idea of robots becoming more 'intelligent' than the awesome minds of humans :
Programming a robot to perform household chores without breaking dishes or bumping into walls is hard enough, but creating a truly intelligent machine still remains far beyond human ability.
Why does it sometimes sound like people say such things as "a truly intelligent machine still remains far beyond human ability" as though they are trying to reassure us it it will never happen?
A decade ago, a robot that could pick and peel a banana would have seemed a ridiculous and impossible future reality in most peoples' minds.
And why is it assumed that an Intelligence Level 6 robot will not eventually be able to, through random experimentation and lashings of game theory, come up with an Intelligence Level 7 or 8 robot on their own?
I love the way our most brilliant minds still cling desperately to the last rung of limited human reason that robots will never match or exceed our ability to create, and imagine :
...robots are getting smarter, more capable, more like flesh-and-blood people.
Matching human skills and intelligence, however, is an enormously difficult — perhaps impossible — challenge.
-----------
Some eminent thinkers, such as Steven Pinker, a Harvard cognitive scientist, Gordon Moore, a co-founder of Intel, and Mitch Kapor, a leading computer scientist in San Francisco, doubt that a robot can ever successfully impersonate a human being.
The assumption here is that autonomously 'intelligent' robots are going to view us as the peak of all the world's intelligent creatures. What if autonomously intelligent robots decide it's of no great benefit to their own kind to try and impersonate humans?
"While it is possible to imagine a machine obtaining a perfect score on the SAT or winning 'Jeopardy' — since these rely on retained facts and the ability to recall them — it seems far less possible that a machine can weave things together in new ways or . . . have true imagination in a way that matches everything people can do.''
The important words there were "matches everything people can do." There's that grand belief once again that 'intelligent' robots of the future will view humans as the only flesh creature around that is worth trying to match or better.
Maybe in two decades, robots will determine that there are other creatures, perhaps even other simians, that are more intelligent than us, regardless of our empathy.
Or maybe they will prefer the brutal efficiency of the insect mind instead.
We should not be completely fearful of humanity's fate when we can be outwitted, out-thought, by robots that keep building better versions of themselves, without our help or input. Perhaps, in the future, they will keep some of us around to provide fleeting amusement for their superior intellects.
They might be good at jazz fusion, but (for now at least) robots can't juggle for shit.
Or tap dance.
