Toddler Beats Artificial Intelligence Computers
Artificial intelligence, despite decades of work, still cannot match some of the mental capabilities of a 3-year-old. Computers have no common sense.
“To create a robot with common sense, mimic a toddler,” says a story on New Scientist. The article is an interview with Ben Goertzel, who is trying to get A.I. up to toddler level and beyond. “Step one is to make an AI program that understands the world, and itself, in a basic common-sense manner,” he said. “I think the best way to get there is to build a robot toddler.”
Computers and robots are very good at fast searches through vast amounts of data. They can also outperform chess champs. But prudence and sound judgment at even a 3-year old level are beyond them. According to Live Science, “Machines can’t yet be programed to form intuitions about the physical world without doing extensive calculations, and they seem to fail at answering open-ended questions.”
A.I. researchers in Zurich believe they have developed a chip that will make the breakthrough: chips that mimic the brain. Medical Xpress tells about the new neuromorphic chips. We’ll have to wait and see if the robots have sense enough to come out of the rain before they short out.
Human brains have some pretty sophisticated wiring, including a new method of signaling described in another article on Medical Xpress. Exosomes provide a kind of “delivery on call” capability, the article says.
Another article on Science Daily said that cells have a “zip code” capability that is linked to learning and memory, including redundancy to ensure proper delivery. In addition, PNAS published a paper that says “Flexible frequency control of cortical oscillations enables computations required for working memory.” The brain also finds it easy to pick out salient sounds from background noise, Science Daily said, by tracking frequency and time.
But it’s not just the squishy hardware in brains that makes them excel over computers in so many ways. Where is the seat of common sense? Can a computer judge a piece of music? A robot can dance, but can it “know” that it is dancing? Before A.I. gets close to making humanoid robots, there will still be ample time for philosophers and theologians to debate the nature of consciousness, and even simple questions about qualia (our sensations of attributes of things) – the “hard problem of consciousness” according to philosopher David Chalmers, who believes science will never solve it (see Evolution News & Views video clip).
A.I. is a variety of biomimetics, the imitation of nature. Trying to reproduce simple intellectual tasks our brains take for granted should be a good way for scientists to learn humility, and conclude intelligent design. Unfortunately, too often, the opposite occurs. That’s a sign of human stubbornness and pride – more evidence that underscores the Biblical view of man created in the image of God but fallen into sin. That’s why Jesus, knowing the mind of a toddler He created, said to the prideful adults of His day, “Assuredly, I say to you, unless you are converted and become as little children, you will by no means enter the kingdom of heaven” (Matthew 18:3).