I think it is a cool idea, but hardly workable.
Unless the AI code is opened (exposing all the evaluation routines) and fully documented (I highly doubt opening the source code will happen)
Still the current implementation is enough to add flavour (altering priorities). The following code fragments are from the FactionDefs file
Code: xml
- <AICategoryWeight>
- <Military>16</Military>
- <Growth>8</Growth>
- <Tech>12</Tech>
- <Diplomacy>20</Diplomacy>
- <Expansion>18</Expansion>
- <Wealth>14</Wealth>
- <Influence>10</Influence>
- <Fortification>6</Fortification>
- </AICategoryWeight>
Terran
Code: xml
- <AICategoryWeight>
- <Military>20</Military>
- <Growth>16</Growth>
- <Tech>14</Tech>
- <Diplomacy>6</Diplomacy>
- <Expansion>18</Expansion>
- <Wealth>10</Wealth>
- <Influence>8</Influence>
- <Fortification>12</Fortification>
- </AICategoryWeight>
Drengin
Still I would love to see more levers, for instance Biases (I have a bias for certain weapons)
And some exposed behaviors (If at war, likelihood to pirate trade routes, likelihood of using planet destroyers...)