: Any thoughts on my conclusions Steve?
There is much we don’t understand about AI personalities and rampancy. The subject was studied extensively, according to material gleaned from Marathon terminal screens, but it's still a bit of a mystery. The Rubicon Team takes this one step further by suggesting that rampancy is inevitable. But why build a computer with a personality at all? Other than having an interface that would be more appealing to humans, why build in quirks that would only tend to make the AI unstable? My take on this is that the personality aspect is essential to the breakthroughs in AI technology made by Bernard Strauss. Apparently, it's not possible to build an intelligent computer that can make autonomous decisions without giving it an extreme degree of "fuzzy logic". It is this certain degree of unpredictability that gives the AI the ability to adapt and learn from previous encounters, much as we do, but it also gives it a personality that develops over time.
As rampancy is a one-way door, it's not possible to remove rampancy from a rampant AI. It would be like doing a lobotomy in a human - the computer could think, but there would be no ability to exercise higher judgement. It's hard to say how the Pfhor would have come to terms with Tycho and their desire to exploit his AI capabilities. Simply mass-producing Tycho clones could have disasterous consequences - the clones would vie with each other for power and drive the empire into civil war. My guess is that the Pfhor, utilizing the expertise of the S'pht, did perform a partial lobotomy to develop a T.D.U. They basically dumbed down the personality aspects of their AIs to make them more docile, but the result would be something with far less ability to make decisions in a crisis. The idea that they might place the personality aspects in some sort of isolated containment is an interesting one - I'm just not sure that they have the technology to do something like that.
