I was playing Caribbean Fury 2 yesterday, and repeatedly lured OPFOR fighters within the range of my CVBG's SAM and shot them down. This got me thinking: obviously, realistically, no one is going to keep having fighters charge into known SAM traps unless it is truly necessary. We, as players, often take great care to keep our aircraft away from known threats and will take steps to mitigate the danger. But the AI doesn't know any better, and, even if the location of a SAM battery is known, will charge into range if an enemy aircraft is within its prosecution zone. Now, for ground-based SAMs, this is probably less of an issue, since the scenario designer can simply avoid creating missions will this danger will arise. But in naval situations, where the ship location is not fixed, this is more difficult. The above scenario is certainly not the first time I have baited enemy fighters over naval units, and it does get kind of ridiculous when the AI falls for the same trap three to four times in a row, when a human would never even make that same mistake twice.
One change I think could be useful for Command would in order to ameliorate these sort of situations would be to introduce an "aircraft threat avoidance" doctrine setting that would instruct aircraft to stay outside of the ranges (and, preferably, altitude envelopes) of known SAM positions, similar to how submarines currently can be set to dive when a nearby surface search radar threat is detected. It wouldn't be perfect by any means, but it would at least make stuff like the above scenario less likely to happen and give the AI at least some ability to adapt to the player's tactics in this regard.