In “The Iliad,” tools in the workshop of the skillful god Hephaestus include voice-activated bellows that seem to know just how hard to blow. Walking with difficulty, he’s aided by golden attendants, “in appearance like living young women,” endowed with intelligence, speech and strength. Self-driving tripods for transportation are among his inventions as well. Journal Report
More in Artificial Intelligence
These are Homeric examples not just of automation, but of machines that seem to know what they are doing. That is in keeping with the definition offered by the AI pioneer John McCarthy, who coined the term “artificial intelligence” in 1955 to mean the making of intelligent machines. Nils Nilsson, another AI expert, said intelligence is what “enables an entity to function appropriately and with foresight in its environment.”
We seem to be enveloped in AI, at least in the news, if not yet in our lives—and there is surely more to come. But how much do you know about its history? Test yourself by taking this quiz—without any help from AI, please.
1. In 1942, in a short story by Isaac Asimov called “Runaround,” a character enumerates three laws governing the behavior of robots. Which of these wasn’t among them?
A. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
B. A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
C. In time of war a robot can take human lives but only in accord with human rules of warfare.
D. A robot must protect its own existence as long as such protection does not conflict with the First and Second Laws.
Answer: C The robot rules in “Runaround” made no exceptions for times of war.
2. In 1950 the English mathematician Alan Turing published a paper asking, “Can machines think?” How did he propose to find out?
A. By having a computer take the SATs
B. By putting a computer in text dialogue with humans. If the humans can’t tell they’re conversing with a machine, that machine can fairly be said to think.
C. By asking the computer to read emotion in a series of human faces. If it can do so as reliably as humans can, it is thinking.
D. By asking the computer to create something, such as a drawing or a piece of music, within parameters set by a person.
Answer: B. The Loebner Prize is awarded annually for the most human-seeming computer program, but nobody has yet won the $100,000 Grand Prize offered for computer behavior indistinguishable from that of a human on a more difficult test involving an as-yet undefined audio/visual component. No computer has come close enough even to require the details of this A/V component to be worked out.
3. Possibly the first successful AI program ran on a computer at England’s University of Manchester in 1952, and was written by Christopher Strachey, nephew of Lytton, the biographer and member of the Bloomsbury set. What did the program do?
A. Play checkers
B. Play chess
C. Play tic-tac-toe
D. Play saxophone
Answer: A. Strachey wrote the program in 1951 for playing checkers, or what the British call draughts, and got a revised version to work the following year, according to scholar Jack Copeland, author of “Artificial Intelligence: a Philosophical Introduction.”
4. In 1956, researchers from various fields convened in a historic Summer Research Project on Artificial Intelligence—at which institution?
A. Harvard University
B. Princeton University
C. University of Pennsylvania
D. Dartmouth College
Answer: D. The Dartmouth conference hosted by Marvin Minsky and John McCarthy is “considered to be the official beginning of serious work in artificial intelligence,” according to the book “The Quest For Artificial Intelligence” by Nils Nilsson. McCarthy coined the term in connection with this conference, writing that, “For the present purpose the artificial intelligence problem is taken to be that of making a machine behave in ways that would be called intelligent if a human were so behaving.”
5. In 1956, AI pioneers Allen Newell, J.C. Shaw and Herbert Simon demonstrated the Logic Theorist, perhaps the first program to simulate humans’ ability to solve complex problems. What did it actually do?
A. It proved 38 theorems by two famous mathematicians
B. It calculated trajectories for intercontinental missiles
C. It solved the traveling salesman problem
D. It enabled an early version of Monte Carlo simulation
Answer: A. The Logic Theorist was able to prove some of the theorems in Alfred North Whitehead and Bertrand Russell’s famous work “Principia Mathematica.” Simon went on to win the Nobel Prize in economics.
6. In the early 1970s, Edward Shortliffe created perhaps the first expert AI system. It was a rules-based program for decision making—in what realm?
A. Tax preparation
B. Medical care
C. Criminal law
Answer: B. The program was called MYCIN and its aim was choosing the right antibiotic for patients with infections.
7. In 1979, the first computer-controlled, autonomous vehicle navigated a room full of chairs. Where?
A. The Bell Labs complex in Holmdel, N.J.
B. IBM’s Thomas J. Watson Research Center in Yorktown Heights, N.Y.
C. The Stanford University Artificial Intelligence Laboratory
D. A basement AI lab at Carnegie Mellon University
Answer: C. AI specialist Hans Moravec writes in the book “Autonomous Robot Vehicles” that the slow but reliable “Stanford Cart” made its way “through cluttered spaces, gaining its knowledge of the world entirely from images broadcast by an on-board TV system,” and that it changed its path as it perceived new obstacles.
8. In 1997, an IBM computer called Deep Blue defeated the world’s reigning chess champion—marking the end of human supremacy at the game. Name the human who took the loss.
A. Bobby Fischer
B. Garry Kasparov
C. Magnus Carlsen
D. José Raúl Capablanca
B. Deep Blue won a six-game match 2-1, with three draws. Since then, computers have grown even better at chess.
9. In the late 1990s at the Massachusetts Institute of Technology, Cynthia Breazeal created one of the first robots able to demonstrate emotional responses arising from human interactions. What was that robot called?
Answer: D. Kismet’s emotional qualities are enhanced by its voice, which is that of a young child.
10. Last year, Google unveiled a new AI technology for enabling a computer to make hair salon appointments and the like using natural conversation so “real” that it is hard to tell it isn’t a person. What’s the technology called?
A. Google Duplex
B. Google Secretary
C. Google Gobot
D. Google Hal
Answer: A. Google Duplex can carry out certain conversations so effectively that the company decided it should disclose at the start of each call that the caller isn’t an actual human.
Mr. Akst is a writer in New York’s Hudson Valley. He can be reached at [email protected] .