AI in Games for Game Developers
AI in games covers two areas: the behaviour of non-player characters, using finite state machines, behaviour trees, pathfinding with A* and utility systems, and the use of generative AI tools in production for concepts, textures, code help or dialogue, with human review and licensing care.
Stage: Optimisation and Release · Part of the Game Developer roadmap · Last reviewed 19 November 2026
What to learn in AI in Games
- Implement a finite state machine for patrol, chase and attack
- Build a behaviour tree with selectors and sequences
- Find a path across a grid with A* and a navmesh
- Pick between state machines, behaviour trees and utility AI
- Check generated assets for quality and licence suitability
AI in Games interview questions
What is the difference between a finite state machine and a behaviour tree?
A state machine has explicit states and transitions, simple but harder to manage as states multiply. A behaviour tree arranges tasks in a hierarchy of selectors and sequences evaluated each tick, making complex behaviours easier to reuse and extend.
How does A* pathfinding work?
A* explores a graph, picking the node with the lowest cost so far plus a heuristic estimate of the distance to the goal. With an admissible heuristic such as straight-line distance it finds a shortest path efficiently.
Where AI in Games fits in the Game Developer roadmap
AI in Games belongs to the Optimisation and Release stage (intermediate level), which takes about 5 hours. Polish and ship.
Other units in this stage
Back to the full Game Developer roadmap