Two Types of Androids

It seems there are two broad classes of things people here think of under the name "android" or "gynoid" - and I'm going to try giving them more specific names and see if they catch on.

1. Toolbot: A machine designed and built to satisfy some human desires (whatever those may be - service, sex, companionship, caretaking...). The machine will be designed to look and behave enough like a human being to satisfy those desires as well as possible given technological constraints. (Examples - Star Trek's Andrea, my Nicole and Katherine, Candy, early Asimov robots.)

2. Personbot: A person, that happens to be made of "artificial" components (and may or may not have been built by humans), but with a personality and capabilities that nobody deliberately gave it. (Examples - Data, Nova, Frankenstein)

It makes sense to me to think of the two differently. I would think of a toolbot as a tool (duh), toy, or slave - but I would be hesitant to think of a personbot as such.

The distinction between personbot and toolbot lies largely in its origin - the reason it was built and the relationship it bears to its design. It may be arbitrarily difficult to look at an existing android and its behavior and determine which category it's in, without knowing how it got the way it is.

There seems to be a cultural idea that a toolbot might (or even probably would) "grow" into a personbot (DB's stories, for example [link to new site pending]). But I've never found that very compelling. If I were designing a toolbot, I'd want to make very sure it couldn't become a personbot - I don't want my tool to turn into a person, either on a practical level (it should do what *I* want it to, not what it wants to) or a moral one (I don't want to feel like I should set it free when I spent time or money on it and am getting value from it).

There also seems to be an idea that humanity couldn't possibly build a toolbot right. (Outer Limits Valerie 23 and Mary 25, for example). I find that depressing. I think we'll probably have some failures - but I also think android designers will be smart enough to do a lot of testing before releasing their creations to the public, and build in failsafes so that even if something goes wrong, the android will shut down rather than cause harm.

Finally, I question why personbots would be built. I could see reasons for transforming humans into personbots - medical, life extension, even cosmetic - and I'd argue that doing so doesn't change their status as people. But why, aside from the challenge and being able to say "I did it," would anyone build a new personbot from scratch?

While the clear distinction between toolbots and personbots seems to make sense in any plausible real-world situation, I'm going to admit that one of the things (not the only one, even in a robotic sense) I find interesting in a sexual way is a combination of the two (a "person-toolbot?") - an android with enough legitimate independent personality to be considered a "personbot," but with whom people interact as a toolbot. They're thought of as just particularly realistic toolbots and are as controllable as a toolbot both in principle and in practice, but nevertheless have enough of a point of view to write from, and are capable of wanting things other than what their masters want.

That's the reason I invented the idea of "donated personalities" for my stories (including Michaela) - it was the best excuse I could think of for justifying the creation of something of that description. I don't think they're very realistically plausible, though; but they do make a good story.

For the record - Nicole, Julia, Katherine and Anna are toolbots; Michaela and Becky are person-toolbots, and I don't think I've ever written a true personbot and probably won't.