Agents acting in environments range in complexity from thermostats to companies with multiple goals acting in competitive environments. Here we describe ten dimensions of complexity in the design of intelligent agents. These dimensions may be be considered separately but must be combined to build an intelligent agent. These dimensions define a design space for AI; different points in this space are obtained by varying the values on each dimension.

These dimensions give a coarse division of the design space for intelligent agents. There are many other design choices that must also be made to build an intelligent agent.