Hey, back off

I feel the same way. They tell you things like "I do not like your starbase so close" or "we know you are about to attack" but you can't tell them the same thing. It would be nice if/when they they make gciv 3 to have a way to take to the other civs. Even as in a basic form it would still be great.

In AI speak it would sound like this: "remove your mobile hardware from this player's SOI". I ended up wiping out the Altarians but it was not easy. I had to go to war much earlier than was comfortable. Our military tech was too close. I actually lost some fleets, but I admit it was more fun this way. I'll never play another game on anything less than Tough.

One interesting observation in this game (not sure if this was a quirk in this game only): None of the other Civs researched mass drivers or armor. Each and every one of them went full bore into missiles and point defense. Some gave a minor nod to beam/shields. Once I noticed that, my ships started cutting through them like warm butter. And they never caught on with armor research. None of them.

Another note: The Drengin kick ass! I set them up as a proxy to distract my 3 biggest rivals, fully expecting the Drengin to get slaughtered, but lo and behold they held on, battle after battle. They were in the center of the map with only about 7 planets, fending off invasion armadas from every direction. I was so impressed I made them an ally and will share the galaxy with them.

The basic problem with this is that the AI values certain things in the game, either --, -, 0, + or ++; and this rating determines his relationship to the other AI'S/you. Ultimately this means either War, or the opposite, Alliance. The game is in fact so modelled that most of these factors are aggravating as the game progresses.

But the human player doesn't need such things. You have all freedom that you want, and can declare war upon everybody anytime, so what's the deal in saying an AI his starbase is too near?

Imagine the human player also bound by such a system - you can only declare war when the rating of an AI is "hostile" - that would be a game-breaker.

Maiden, I agree Freedom of Choice is essential in this game. Sometimes I want to extend that same choice to an AI. I want the option to warn an AI his current behavior is about cause us both some inconvenience. The way it is now, I have an almost unfair advantage by virtue of knowing the AI's mood toward me. It would only be fair to extend that courtesy to them as well. However our actions do not need to be locked into those declarative moods by any means. On the other hand, the AI should not be locked in by those moods either (to make it fair). They should be able to launch a surprise attack if it works out to be in their best interests.

The suggestion of "warning" AI (which has been brought up many times before) isn't meant to change the motives of AI (it doesn't change a human's objectives when the AI "warns" them either), but merely to force them to make some stronger efforts to conceal their plans better.  That has benefits (force them to keep their distance) and problems (when they do attack it may be a bigger surprise), but I don't think this requires a change in how the AI prioritizes their real objectives - simply how they go about accomplishing them.  (Or not so "simply", since we're talking about AI programming here...)