These are all non game specific suggestions:

+1 Russell and Norvig, Artificial Intelligence: A Modern Approach

You really can't beat this book for a broad introduction to all things AI. It covers evertyhing from neural nets and hidden markov models, to logic and search.

+1 Tom Mitchell, Machine Learning

My only complaint about this book is it's from 1997 so it's missing some of the new things that have really reshaped machine learning (namely support vector machines).

+1 Christopher Bishop, Neural Networks for Pattern Recognition

Don't get this for a broad introduction, get this if you want to go deep enough into feedforward neural networks to understand and implement them.

To WeirdoFu: I can't say I agree with you. I think instead AI is just becoming more and more grounded in probability, statistics, and learning theory--things like AdaBoost, Support Vector Machines, Kernel have had amazing practical success but are also very nicely backed up by theory.

Quote:
I realize that some people will say that pathfinding is not AI

"Artificial Intelligence is the Science of making machines do things that would require intelligence if done by men."
- Marvin Minsky

I agree with Mr. Minsky, so i'de say pathfinding is AI.

Right now I'm looking primarily at "Artificial Intelligence: A Modern Approach" and "Reinforcement Learning: An Introduction", so I'll probably end up choosing one of those. Both were recommended and after looking through the ToC and browsing a few pages through Amazon, I'm fairly confident I would be able to understand the subject material.

Quote: I agree with Mr. Minsky, so i'de say pathfinding is AI.

What I actually meant to get across was that some people have told me Searching Algorithms (meaning A*, Breadth-First, ...) are not AI by themselves but rather tools to be used in AI (for pathfinding that is).

I usually go with the philosphy that if I can understand the first chapter or so of a book, then I could probably get myself to understand all of it after beginning to read it. Looking at the ToC of "Tom Mitchell, Machine Learning", even the beginning chapters are a little-bit beyond my current level of understanding, so I'm going to avoid it for now. The other 2 books I mentioned above seemed easier for me to comprehend, and I have some knowledge of the topics near the beginning of the ToC.

Whats covered in "Artificial Intelligence: A Modern Approach" is really what I was looking for, but with Amazon returning "3,379 relevant results" for "Artificial Intelligence", I would have had a difficult time finding the book without these recommendations. Thanks.