Debate: Can AI Intelligence be Measured?

  • Thread starter Thread starter Ferno
  • Start date Start date
  • Tags Tags
    Ai Intelligence
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
6 replies · 3K views
Ferno
Messages
4
Reaction score
0
In one of our laboratory's coffee breaks, a debate arose but no one was able to give concrete answers. One of the questions was the following:

What defines a better and a worse AI? Is it simply the speed in which the problems are solved? Or is there a maximum intelligence unrelated to speed?

Is there some evidence of this?

Cheers.
 
Physics news on Phys.org
AI is NOT about speed, it is about simulation of higher brain functioning such as reasoning.
 
The best definition of AI that I've ever heard was stated during a colloquium at Los Alamos:
...Artificial Intelligence is a field in which one attempts to find well-defined algorithms that solve ill-defined problems...
 
Last edited by a moderator:
Ferno said:
But reasoning is also speed related.
No it isn't. No human can beat a handheld calculator for speed at basic math functions, but that doesn't mean a calculator is AI. Computers today are plenty fast enough to beat humans at virtually any pursuit we can figure out how to program them to do. But it is difficult to program them to think like people.

As others have said, AI is almost entirely a matter of programming.
 
I like gdp's quote:
...Artificial Intelligence is a field in which one attempts to find well-defined algorithms that solve ill-defined problems...
Note that this definition doesn't touch on consciousness or self awareness. An automated car able to stay on a roadway and navigate without human intervention might be a good example of an AI system.

AI and strong AI are generally considered 'different' primarily in that Strong AI implies sentience/consciousness.

Definition of AI here:
http://www-formal.stanford.edu/jmc/whatisai/node1.html

John Searle coined the term "strong AI" in a paper here:
http://www.bbsonline.org/documents/a/00/00/04/84/bbs00000484-00/bbs.searle2.html
(albeit, I'm not absolutely sure this is the first paper that references the term "strong AI", but I believe it is)
 
Last edited by a moderator: