A team at the University of Illinois at Chicago said it put ConceptNet 4, an artificial intelligence system developed at the Massachusetts Institute of Technology, through the verbal portions of the Weschsler Preschool and Primary Scale of Intelligence Test, a standard IQ assessment for young children.
ConceptNet 4 has the average IQ of a 4-year-old although the machine's scores were very uneven across different portions of the test, unlike those of the typical child, a university release said Tuesday.
"If a child had scores that varied this much, it might be a symptom that something was wrong," UIC computer scientists Robert Sloan said.
While ConceptNet 4 did very well on a test of vocabulary and on a test of its ability to recognize similarities, Sloan said, it performed "dramatically worse than average on comprehension -- the 'why' questions."
Computer programs frequently have problems making sound and prudent judgments based on a simple perception of the situation or facts, he said -- the dictionary definition of common sense.
Common sense has proved difficult for artificial intelligence designers because it requires both a very large collection of facts and what Sloan termed implicit facts -- things so obvious that we don't know we know them.
"We're still very far from programs with common sense -- AI that can answer comprehension questions with the skill of a child of 8," he said.