Game AI Roundup Week #41 2007

Alex J. Champandard on October 13, 2007

There are 9 more Smart Links from the blogosphere this Saturday. Visit the tumbleblog at (or subscribe by RSS) for updates on game AI as they happen throughout the week.

Avoiding Ten Common Game AI Mistakes

Justin presents 10 different problems in game AI, including movement, animation, oscillation, unpredictable and erratic behaviors… Luckily, most of these aren’t too hard to fix!

Inside Bungie Podcast: Halo 3’s Game AI

Damian Isla discusses the challenges of building AI for a next-generation games, and the process bungie went through to create a fun experience. (Thanks Jeff.)

Podcast: Inside Bungie on 16/08/07 (MP3, 71.6 Mb)

Next-Gen AI at GDC France

Guess who’s giving a talk at GDC in Lyon this December? My topic is behavior trees and next generation AI. I’ll be discussing certain ideas I’ve written about on this blog, but some surprises also… :-) I hope to see some of you there!

Façade in Augmented Reality

Congratulations to the Grand Text Auto team for launching their exhibition, featuring their work in interactive fiction.

Neurotic AI For RTS Games

Austrian researchers created aggressive, defensive, normal and neurotic versions of the AI software in the war strategy game Age of Mythology, drawing on the big five emotional dimensions to personality recognized by psychologists.

Confessions of an Academic Crack Smoker

Julian Togelius sets the record straight in the next episode of this drama…

First Interactive Sitcom?

More Than Interactive Games is trying to revolutionize the sitcom with the launch of the first episode of “Supple®,” a new downloadable dialog-driven interactive game.

The Artificial Intelligence Designer for Video Games

Loki discusses the importance and the role of an AI designer in the game development team. He’ll be quite comforted to know there are already quite a few people in this position in industry already!

Machine Learning and Dragons - a Game

In this game, you’re a knight and your job is to kill as many dragons as you can. The twist is that the dragons use Genetic Programming to learn from every encounter. (You can optionally have them use Reinforcement learning instead too.)

Stay tuned next Saturday for more smart links from around the web.

Discussion 0 Comments

If you'd like to add a comment or question on this page, simply log-in to the site. You can create an account from the sign-up page if necessary... It takes less than a minute!