Artificial intelligence is becoming more and more integrated in our lives. Beyond Siri, A.I. help is woven into apps that help us find restaurants or a nearby gas station. Cleverbot, an online A.I. chat, allows you to have trivial conversations with a robot. So far, A.I. is clever, but hardly intuitive, but that could soon change with the development of ConceptNet.
ConceptNet is the literal brainchild of an open-source computing project at MIT. Not only is the program smart, but also it has the I.Q. of a four-year-old child — the age when children begin to make more complex inferences. If the MIT team is able to build this intelligence, developments in A.I. could take off at rapid pace.
[A]ssessing intelligence is just as complex as intelligence itself. The same could be said of A.I. While ConceptNet may have the I.Q. of a four-year-old child, does it have the E.Q. of one as well?
Elon Musk has repeatedly predicted doomsday scenarios with the escalation of A.I. developments, but before building a bunker to protect yourself from the robot-calypse, there are some intricacies about I.Q. that need to be considered. Take the age-old intelligence debate between cats and dogs. Dog owners will argue that their pets are smarter, and it’s been proven that canines have the I.Q. of a two-year-old child. But cats have twice the neurons of dogs, which means they have the potential to analyze much more complex problems and make more subtle inferences. That said, cats’ brains have not grown in size since they were domesticated 8,000 years ago, while dogs are much more social and thus have brains that have continued to grow. So who’s smarter?
It’s hard to tell because assessing intelligence is just as complex as intelligence itself. The same could be said of A.I. While ConceptNet may have the I.Q. of a four-year-old child, does it have the E.Q. of one as well? Or, to put it in cat and dog terms, does it think like a cat, but have the potential to relate like a dog?
Looking to the animal world not only may help us build better A.I., but also could help us better understand our future relationship to it. Intelligence is tricky — we hardly understand our own and billions have been spent trying to map our brains. Even if mind-mapping becomes possible, then what? Intelligence is rife with paradoxes. How else can it be explained that both pigeons and monkeys can do the same level of abstract math?
Just how does A.I. learn new information? Microsoft Director of Search Stefan Weitz explains:—
Daphne Muller is a New York City-based writer who has written for Salon, Ms. Magazine, The Huffington Post, and reviewed books for ELLE and Publishers Weekly. Most recently, she completed a novel and screenplay. You can follow her on Instagram @daphonay and on Twitter @DaphneEMuller.
Image courtesy of iStock