How bright is the brightest artificial intelligence on the planet?
Well, according to researchers at the University of Illinois at Chicago, it's somewhere around the level of a four- year- old.
Testing ConceptNet 4, an MIT-developed artificial mind, scientists pegged it somewhere in that neighbourhood, using the Wechsler Preschool and Primary Scale of Intelligence Test.
A standard IQ testing tool for children between the ages two and seven, the Wechsler system consists of three components -- verbal, performance and full-scale IQ. Illinois researchers tested ConceptNet 4 only on the former component.
According to University of Chicago at Illinois News Center, the test results were widely uneven.
“If a child had scores that varied this much, it might be a symptom that something was wrong,” Professor Robert Sloan, lead author on the study, explained in the report.
While the AI sailed through the vocabulary aspect of the test, it stumbled on comprehension -- the so-called 'why' questions.
"One of the hardest problems in building an artificial intelligence is devising a computer program that can make sound and prudent judgment based on a simple perception of the situation or facts -- the dictionary definition of commonsense," Sloan explained.
It may seem a long way off from plunging the Earth into a nuclear holocaust and then building near-indestructible robots to hunt down humanity's last stragglers (and maybe going back in time to tie up a few loose ends), but ConceptNet 4's results are still impressive.
The key takeaway, Joe McKendrick writes in SmartPlanet, "is that AI — which is increasingly being adopted into decisioning systems in areas such as healthcare and financial services — is doing a good job retaining and spitting out data to answer natural-language queries. But it’s going to take some time before these systems come close to figuring out the way the world works."
After all, at some point, even Hal 9000 had to Hal... 4.
Also on HuffPost