Robotic Capabilities (1).

General chat about fembots, technosexual culture or any other ASFR related topics that do not fit into the other categories below.
Post Reply
User avatar
fembot_stalker
Posts: 117
Joined: Thu Feb 02, 2012 11:34 am
Technosexuality: Built and Transformation
Identification: Human
Gender: Male
Location: Richmond, Virginia USA
x 21
x 3
Contact:

Re: Robotic Capabilities (1).

Post by fembot_stalker » Wed Apr 05, 2017 12:34 pm

daphne wrote:The Three Laws, I don't mind saying, are garbage.

The main problem with them is that Asimov stated them in such a way as to offer a simple, mechanized way of codifying morality, but the language is inherently interpretive.

For example: "A robot may not injure a human being or, through inaction, allow a human being to come to harm."

Okay. Suppose two humans are fighting? Which one do you help? There's only one of you, so you can't stop both of them, but by curtailing one, you give the other the advantage.

Suppose a surgeon is cutting into human flesh. Technically he is injuring a human being. Do you stop the procedure or not?

How far does "harm" go as a definition? Is a heroin junkie harming themselves? Is a masturbator? What about someone who's depressed over losing a loved one? How would you even keep them from doing that?

And so on. Like many classic era sci-fi writers Asimov falls into the trap of technological solutionism. You can't innovate your way out of, or into, morality.
Saya wrote:Believe it or not, Daphne (and not to derail this topic into a discussion about the philosophies of Asimov) that was actually intended by Asimov. The idea of "Allow No Human To Come To Harm" being ill-defined appears in a lot of his works, but most prominently in the form of the "unlisted" "Zeroith" law, which is when the robot comes to the realization that they more or less have to enslave humanity to keep it from killing itself.

It's used in other ways, too, like a short story in which a positronic computer on a starship encounters a conflict with its first law directives, because the jump drive the ship uses causes the crew to "die" or get knocked out of existence for a fraction of a second, which causes all sorts of puzzling errant behaviors. And then there are times like in "Runaround", where the first law is less of a hazard and more of an active hindrance.

In short, this is just me trying to stick up for Asimov, because a lot of people tend to knock them as being stupid and flawed when being stupid and flawed was sort of the point from the start, because if it worked perfectly, what conflicts would Calvin, Donovan and Powell need to solve to make for a good story?
Murotsu wrote:Here's my initial list of requirements. Like Saya, I see no real reason you'd want it to obey Asimov's Three Laws. These leave far too many holes that can be exploited. For example, you could tell a robot under those laws to commit a crime so long as a human wasn't directly involved.

Anyway, here's my list of basic qualities you'd have to meet for a humanoid type robot:

Weight: Approximately equal to a human's. A much heavier robot creates all sorts of problems from needing more power to inertia issues, to structural ones using equipment meant for humans.

Size: Approximately the same as a human. Ideally, it would be the same in physical dimensions.

Sensory capacity is sufficient for it to have near human qualities in optical, auditory, and tactical areas.
** Optionally it might include olfactory and taste but these are not necessary for a basic humanoid robot to interact normally.
** Additional sensors as required to perform human-like functions where a robot would require them even though a human doesn't have them.

Has a equal to or greater than human flexibility, range of motion, and sense of balance.

Is not in the "uncanny valley" when in common or widespread use.
This differs from a one-off unit or a rare robot in that humans are used to seeing them as presented and not concerned or surprised by their appearance.

If it is to pass for human it must at a minimum do so when casually viewed.
That is, if you passed one walking on the street you wouldn't pay any particular attention to it.

Strength is equal to or greater than a human's. Being super strong is unnecessary as a baseline.

Ability to interact with humans verbally and through body language or facial expressions. The degree required doesn't have to be 100%, simply sufficient that a human would understand the expression or movement correctly.

Reliable and or easily repaired / maintained.
Both would be better but one or the other is a necessity.

Upgradable.
This is important to allow for longer service life and to keep it current with new technologies.

Has sufficient internal power for a reasonable length of operation
I'd say 8 to 16 hours would be desirable.

Can be recharged / refueled in a reasonable length of time.
I'd put the maximum at about one hour. The time is dependent on length of operation between charges / refuelings.

The power source / fuel is commonly and widely available.
This means the robot can be recharged or refueled almost anywhere it goes.

Damage resistant to normal daily hazards.

The computer / processor, memory, data, and programming is sufficient to allow interaction with a human or another robot to carry out all the intended end use functions without human intervention. The robot would only require general instructions or directions.

Ability to pass a very advanced Turning Test variant.
That is, the robot can't be easily confused or fooled by "trick" answers. It also isn't simply doing things by rote or following a script. The programming has sufficient flexibility to allow the unit to make decisions, logical choices, and be rational in them.

Has a set of moral values in its programming.
That is, in the broadest sense it does the "right thing" in almost any situation. Example: You couldn't tell one to knowingly commit a crime.

Is safe to use in terms of chemicals and materials it is made from.

Now, from this you could add additional features to make a robot into more of one that suits your individual needs.
Stephaniebot wrote:If they ever discover how to download a human mind, into a robot body... ;)
Murotsu wrote:
Stephaniebot wrote:If they ever discover how to download a human mind, into a robot body... ;)
Well, that's coming at some point. You really don't even have to download (upload?) the entire brain, just personal memories, personality, emotions, and other data that is unique to the person. Most of what your brain does is stuff that could be considered universal in nature.

Take for example, facial recognition. There are already programs that will do this.

In other cases, there are things that as a human you would do, but probably wouldn't have to as a robot / droid. An example of this might be getting an itch or coughing. If such behavior was still deemed desirable and necessary a canned "fidget" program could be substituted.
Stephaniebot wrote:It will come at some point, for sure, whether it happens soon enough for me to have that delight done to me (I'm 59 next month) is another matter entirely, but fingers are crossed. And yes, you're right, there would be no need to download everything, but working out where the critical data is, and sifting it out from the rest might be a challenge. Indeed, for me, I wouldn't want to be 'too human' anyway!


Wow! You all are so incredible! You have devised all types of theories about robotics and AI in relation to IA's 3 Laws that I would have never even thought of before. I think at that time when IA first drew up his 3 Laws, he wanted them to be simple enuff for the average person to understand - especially given the fact that computers and computer chips were not as nearly as sophisticated as they are now. Also, I thanks for the extra list of capabilities too; I would definitely use some of those for my (female) robot(s) as well. Here are some more capabilities that I have decided to use, even though they have already been thought up:

39. Energon/Electromagnetic wave detection and music identification software such as Spotify/Shazam
40. Highly sophisticated IRMs (internal repair mechanisms) - and nanotech utilities that enable any robot to be able to control other machines.
(alt.sex.fetish.robots) AI and UFOs are spreading all over the Globe! 2 see more, just visit us at Malestrom1000's Youtube Channel. Thank you 4 visiting us as we get closer 2 creating our (female) repairbots, (giant) r⊙b⊙t gladiatrices, fembots, sexbots & h⊙b⊙ts; now how pimpadelic is that? :transformer: :rockon: :thumbsup: :nerd: :D :) :-o 8) :lol:

Post Reply
Users browsing this forum: No registered users and 20 guests