I’m starting to think more about how constructs are constructed for a purpose, and how this is expressed in Murderbot. constructs have a set of core priorities. “instincts” if you will. and this seems to be across the board, from constructs to bots (including high level bots like ART) to systems.
MIs and constructs have a purpose, and that purpose is usually people-oriented in some way shape or form.
along this thread, i believe pack-bonding, or a coded affinity for social ties, is a built-in failsafe for most machine intelligences. basically, if you make the powerful sentient machine care about its people, make it social, it’s less likely to turn on them.
what I’m getting at is that Murderbot’s self-destruction is something ingrained. it can’t really fully see or come to terms with its self-destruction because its function *is getting hurt (and even dying) for clients*.
The “this is how I win” threads in Exit Strategy and Network Effect come to mind. the ultimate “winning” or satisfaction or purpose, in Murderbot’s mind, is to die killing a particularly nasty threat. To sacrifice itself. That Murderbot2.0 says this is especially damning wrt a programmed purpose.
I also think about the moment in Network Effect when Murderbot says it's guilty about having Thiago and Overse help it with a plan. To its mind, it has no worth if it’s not fulfilling its function. I mean, its guilt about this starts with Overse and Thiago insisting on accompanying MB in checking out the space dock: MB pushes back, but then it finds out Pin-Lee also wanted it to be safe by including that stipulation in its contract, and then the cherry on top is ART telling Murderbot that it can't lose it. It matters to all of these beings that MB remain whole and alive. and Murderbot is *so* confused by this. It literally says, “it was confusing. I was confused.” Murderbot's genuine confusion about being cared for by all these people is overwhelming to it, it literally doesn’t make sense on a fundamental level.
like the implications that it’s loved and important to these people—that its function is secondary to its personhood, that its people would rather be in a little more danger than have it endanger itself unduly—is *really* hard for it to reconcile.
like! this can’t be overstated. it was created to protect people, and self-sacrifice is so deeply a part of it that it’s very possibly in its *code*. i mean, even if it’s not literally coded into it, its experiences (its early treatment), have shaped its beliefs around what its function is. it’s behaviors.
like how do you deprogram someone from being self-destructive when its literal purpose is self-destruction?
