Mind your NPCs and Q’s: what’s next for Game AI?
summary
Overview of concrete work and demos not platitudes: middleware Kythera is presented with live features — ORCA+ for dynamic avoidance, runtime nav links and auto cover generation, an “interesting path” system used in Wilson Lords of Mayhem, and a city-scale simulator that spawns pedestrians and vehicles. Discussion covers algorithmic details (A*, Jump Point Search, Anya, hierarchical pathfinding, context steering), spatial query and influence maps, modular behavior pipelines, and speculative uses of ML and GPT-3 style synthesis for dialogue and procedural quests.
AI-written summary · about
chapters (23)
- 0:00Game AI introduction
- 4:54Defining game AI and its role
- 7:03Pathfinding in games
- 11:15Steering algorithms
- 14:24Motion and animation
- 16:53Creating interesting paths for NPCs
- 17:35Dynamic avoidance
- 18:19Spatial reasoning
- 20:27Spatial queries
- 23:59AI decision-making
- 25:44Future decision systems
- 27:10ML and authorial control
- 27:52Meta AI level gen
- 29:37AI directors
- 30:40Companion AI
- 33:51Living world challenges
- 36:40World building
- 37:45AI embodiment
- 38:46Shared AI libraries
- 40:30ML in game AI
- 41:12Future game AI trends
- 42:34Intelligent game worlds
- 44:24Call to action
related talks
59mGame AI 4N00bz
52mThe Journey of AI in Games: A Technical and Philosophical overview
19mDynamic storytelling and NPC interactions using generative AI & Azure Cosmos DB in game development
01h 01mIntroduction to Practical AI
32m..Stories About Industrializing In-Game Machine Learning Solutions
39mBuilding self evolving video game AI in Python
29mThe Present & Future of AI and Games
14mProcedural Simulation, AI creativity and stochastic modeling. Using Azure Cosmos DB to make games.