AI is concerned with the development of autonomous agents (computer systems that sense and act). (*)
A quote from Genesereth & Nilsson [Logical Foundations of Artificial Intelligence 1987] page 6:
The view taken toward AI in this book follows the theme hinted at by Leibniz and Frege and then substantially elaborated and developed into specific proposals by McCarthy [Programs with Common Sense 1958]...
It is based on two related ideas:
- First, the knowledge needed by intelligent programs can be expressed as declarative sentences in a form that is more or less independent of the uses to which that knowledge might later be put.
- Second, the reasoning performed by intelligent programs involves logical operations on these sentences.
For a more up-to-date overview see the landmark volume by Jack Minker (editor): Logic-Based Artificial Intelligence 
(*) In principle such agents range from simple thermostats over advanced chess programs to robots that may eventually pass the Turing Test.
Jørgen Villadsen 2006-12-30 imm.dtu.dk/~jv/ai