The Turing Test

Event
  • Add to Calendar 2023-10-31 15:00:00 2023-10-31 16:00:00 The Turing Test As part of the rules for the Turing Test (AKA “the imitation game”), Alan Turing proposed in 1950 that a machine can be said to be intelligent if it has the ability to exhibit intelligent behavior that is indistinguishable from that of a human. Though a controversial metric for artificial general intelligence, a prize has been offered every year since 1990 for any system that can fool a significant number of judges into believing that a human is on the other side of chat-based conversation. How does the Turing Test hold up at a time when AI has entered into new areas of creative human activity?<br /> <br /> Topics of conversation might include: Is the ability to parrot human behavior or use of language a true test of machine intelligence? Is the Turing Test more or less relevant with the emergence of machine learning, large language models, and text-to-image generators? What are other limitations of the Turing Test as a true measure of intelligence? What are better alternatives? 03:00 pm 04:00 pm America/Los_Angeles public
October 31, 2023 - October 31, 2023
3pm
Ahmanson Lab | Leavey Library, 3rd floor (LVL 301)

turing test
turing test

As part of the rules for the Turing Test (AKA “the imitation game”), Alan Turing proposed in 1950 that a machine can be said to be intelligent if it has the ability to exhibit intelligent behavior that is indistinguishable from that of a human. Though a controversial metric for artificial general intelligence, a prize has been offered every year since 1990 for any system that can fool a significant number of judges into believing that a human is on the other side of chat-based conversation. How does the Turing Test hold up at a time when AI has entered into new areas of creative human activity?

Topics of conversation might include: Is the ability to parrot human behavior or use of language a true test of machine intelligence? Is the Turing Test more or less relevant with the emergence of machine learning, large language models, and text-to-image generators? What are other limitations of the Turing Test as a true measure of intelligence? What are better alternatives?