When the time comes — and you know it will — would you marry your robot? If yes, why? If not, why not?



Here’s the research suggesting we’re soon going to be loving metal and plastic more than blood and flesh:

David Levy, a British artificial intelligence researcher…
wrote in his thesis, “Intimate Relationships with Artificial Partners,”
that trends in robotics and shifting attitudes on marriage are likely
to result in sophisticated robots that will eventually be seen as
suitable marriage partners….

The thesis examines human attitudes toward affection, love and
sexuality and concluded that the findings are just as applicable to
human interaction with robots of the future as they are to the
relationships between humans of today.

If you are a man — do you currently have any sexual experience with
soft plastic? A blow up doll? A rubber vagina?
If you are a woman — do you have any present sexual experience with a
vibrator? Plastic penises? Other mechanical devices?
If we seek private sexual pleasure in the mechanical, why would we not
seek public psychic sorrow in marriage… to a robot?

Is the leap from obedient machine and submissive rubber to cogent robot
and artificial intelligence one too hard to ponder and a conceit too
impossible to consider?
Would you vote for a robot president?
Would you accept a robot teacher in the classroom?
Would you allow a robot physician to operate on you?
Would you want a robot judge in the courtroom?

42 Comments

  3. Scott!
    Why does the Turing Test matter? How many people could pass that test? 😀
    Do you know how many men would love an obedient, “No Questions Asked” robot wife?

  4. Nicola —
    I’ve heard women say how much they’d love to marry their microwave… does that count as a robot? 😀
    I think there will be many more “robot mistresses” than robot wives… though robot wives might not want half of community assets in a divorce… but then they would probably want full custody of all the motor oil. :mrgreen:

  6. Would you vote for a robot president?
    Not unless it passed a Turing test.
    Would you accept a robot teacher in the classroom?
    I think a robot would be good as an adjunct to a biological teacher.
    Would you allow a robot physician to operate on you?
    Yes! I think it would be my preference.
    Would you want a robot judge in the courtroom?
    Yes! Robot juries too. Remove the human element from the courts system.

  7. jonolan!
    I think we’re putting too much preciousness on the Turing test! You can cheat on any exam! 😀
    A robot adjunct would be funny. Make the ‘bot read all the term papers and correct all the grammar.
    In some ways we already have robot surgery — remote operations where “people” control robotics to operate on people in different countries.
    If we have robots do we need courts? Wouldn’t we just press information on the case into a computer and let the “robot brain” decide guilt and innocence?
    I wonder how a robot would determine pain and suffering? Would they use actuarial tables?

  8. Don’t we have a robot president now? He just does everything he is programmed to do.
    A robot surgeon would be good because surgery is a purely objective science. The response to a patient going into cardiac arrest mid-surgery would be immediate and the surgeon might even have those paddles as an attachment. 🙂
    Trials, on the other hand, are very subjective. When you are dealing with things like perspective (“You say you saw the man stabbing him, but you were looking at a rather obtuse angle…”) and the state of a person’s mental facilities, only a human could really get to the bottom of what is going on inside another human.
    Would a robot teacher be able to spot the change in a persons face in the classroom indicating that someone just didn’t get it? Several faces at the same time? Most likely not. A robot professor would not have any genuine interest in the subject matter and that lack of real interest would be patently obvious.

  9. As far as the judicial system goes, I’d love to have something like computers or robots decide everything.
    Picture a court case where no mention no mention of: race, religion, ethnicity, education, etc… were mentioned – only the facts of the case.
    Is the defendant a poorly educated urban Black or a White suburbanite? There would no longer be anyone with decision making power who would care!
    Yeah! Computers or robots for the courts!

  10. At the moment the only robot I really care about is the one that made this information appear on my screen : 7:13 AM On FedEx vehicle for delivery SEATTLE, WA

  11. David,
    LOL!!! I actually told someone on a blog once that they were failing the Turing Test.
    Many people may not find the ability to pass the Turing Test a requirement for either a short term or long term sex partner. I’m of the rare breed that believes that the sexiest organ in the human body is the brain. I would not be interested in someone who would fail the Turing Test.

  12. Gordon!
    I think we have a robotic president, yes: Dark skinned people are bad; old, white-haired, pale-skinned conservatives are good. He will not believe what Putin says to the press about Iran without privately asking him first — but he will wholly believe everything Iran says to the press without ever wanting to get in touch off-the-record! Who knew robots were hypocrites?
    How would your robot surgeon determine how much cancerous tissue to remove? Isn’t there a human judgment factor involved where taking too much or too little is an ongoing determination when it comes to needless suffering?
    I wonder which way a robot would think our dancer from yesterday is spinning?
    I agree a robot would not be good at babysitting a class — but robot grading would be grand because no favoritism would be injected into the pursuit of the grade!

  13. jonolan!
    But what about human significance? Empathy? Determination of punitive damages?
    What if Race and ethnicity and gender were part of the case? How does a robot quantify the terror of a noose on a door handle?
    How does a robot judge if a woman was sexually harassed at work or not based on a “hostile work environment” claim?
    Doesn’t the law require a modicum of street sense and human empathy to dispense justice?

  15. Gordon —
    I can’t believe your robot didn’t arrive yesterday! I think you’re the first person I know who didn’t beat the drop dead delivery by a day.
    Let us know when you’re married! 😉

  16. I believe that under the Rule of Law there is no human significance in the courtroom. Punitive damages would be dealt with through a codified system of values built into the Law. Read up on old Anglo-Saxon laws involving bloodprice (wereguild ) or old celtic Brehon Laws.
    As for Race or Gender being part of the case, I can’t see where they would be. I do NOT accept the legal propriety of “hate crime” legislation. Actions are important; motive only speaks to whether the crime was premeditated or not.
    I admit the sexual harassment case would be difficult to work with though.
    As to “street sense” or “human empathy – no, I do not think they should be components in the criminal “justice” system. I believe the law should be a mechanism.

  19. jonolan!
    How are your robots determining the subtle shift in values over time?
    Is there a cutoff date — say April 21 to April 22 — where sex outside of marriage is not as bad as it used to be in a divorce case, or same sex sexual encounters do not destroy the fabric of the community, or that prostitution and alcohol consumption are tolerable in certain communities when regulated and taxed?
    How will your robot judiciary deal with the appeals claiming the ongoing lack of a legal blind eye: On one day you decided this was legal and on another day you changed your mind because of a shift in community standards?

  20. Shifts in values would be handled by revisions to the tables. That would be a human intervention as would determining what those values actually are in the tables.
    Dates are dates. If the “crime” happened when it was a crime then apply the standard. if not, don’t.

  22. I’m thinking a computerized / robotic judge to do preliminary checks on the Prosecution’s and Defense’s inputs and a separate jury subsystem to render verdict and sentencing.

  23. jonolan & David,
    Wouldn’t time just be a weighted factor for the computer when comparing against related precedent? For statutory law, I’ve long been of the opinion that all laws should have a predetermined expiration date. If they’re still good on that date, they’ll be renewed. I’d say a max of 25 years on any law should be enough. Certainly no more than 50.
    One serious question regarding your robot jury though, and actually, for humans as well, what about bugs in the system? Instead of each case being different, we could end up with 100,000 cases all suddenly being overturned because a bug was found that caused all of the convictions.

  24. LOL! Obviously it wouldn’t be a system designed by Microsoft! Truthfully it would need to be a very robust system with lots of in-built checks to make sure that bugs didn’t cause havoc.

  25. jonolan —
    So your robot judiciary would need to have sentencing guidelines for every infraction, right? Does that include probation and fines?

  26. Scott!
    I think a computer glitch is the fairest way to determine guilt and innocence: If there’s a glitch, let them all go or retry them if you can. It’s better to let everyone reboot then only a selective reboot that requires various degrees of danger and judgment to decide who was wrongly punished in the glitch or not.

  29. David,
    Are you spec’ing this thing out or something? Just remember “the original art” is mine.
    Yes, age alone. White lines and black letters all around.

  30. jonolan —
    Yes, I’m creating my own robot arbitration service. I’d like you to be Chief Justice of this new venture, but we have a little implanting to do first…
    Speaking of implants… how would your robot judiciary handle the Supreme Court? All robots? Half robot half vibrator? How will those interpretive decisions be made? Will the strict constructionists win? Is the Constitution meant to be a living, breathing, document or not?

  31. The robot wife has arrived. Now I just need to let her charge fully. Uhm, and then unplug her, use her until she nearly goes to ‘sleep’, turn her off, and then plug her in again? Something about battery conditioning. Whatever it takes to avoid having a battery that lasts 20 minutes like my present laptop does – I plugged it in and turned it on right when I got it and the battery has been rubbish ever since!

  32. Hi Gordon!
    You aren’t supposed to have to baby the battery like that with the new machines. If you feel like doing it, great, if you want to get on with your life, then do it!
    Congrats on the new spouse! 😀

  33. Well, from an objective perspective, I wouldn’t think there would be any difference between a doll/vibrator and a AI equipped robot.
    Ultimately I think it would all boil down to increasing the level of complexity to meet the needs of the task at hand. I think most of these things are real possibilities.
    With the current work being done with intelligent interactive robotic systems, facial expression/recognition, etc, I’m actually fairly certain that AI analogs could eventually be found for things like teaching and surgery.
    Marriage? Meh. Why would you want to? Unless they became self aware and were able to truly interact with you on an emotional level, it wouldn’t be much different from marrying your vibrator. At which point they may not want to marry you anyway. More power to anyone who wanted to, but no point IMHO.
    So far as AI lawyers, judges and CICs, actuarial tables might work for simple cases, and I agree with Jonolan in saying that even punitive damages, and fines could be meted out that way and still work.
    However the one big problem I see with actuarially based AI justice is not the penalties, but rather how the judgment itself is made. There are many cases that would not fit neatly into a table based evaluation. Also the system could be abused, since a case could intentionally be constructed to trigger specific responses.
    Statistics is good for generalization, but make for horrible individual assessment. You would still require human intervention to make sure this was not the case. Which would make the AI an unnecessary overhead.
    You also have to wonder, If we ever developed self aware AI equal to our own, what guarantee is there than they won’t be equally quirky?

  35. Ew!
    THE ROBOT TODDLER LIVES!

    OSAKA, Japan – A group of scientists in Japan have developed a robot that acts like a toddler to better understand child development.
    The Child-Robot with Biomimetic Body, or CB2, was developed by a team of researchers at Osaka University in western Japan and is designed to move just like a real child between 1 and 3 years old.
    CB2, at just over 4 feet tall and weighing 73 pounds, changes facial expressions and can rock back and forth.

    http://www.msnbc.msn.com/id/19112210/

  37. David,
    what a fascinating discussion!
    i agree with you that the turing test is no measure of humanness or suchlike intelligence. we can take AI seriously when a robot refuses to take the test and provides a compelling defense of it’s stand!
    isn’t it also interesting that the quest to make robots more like humans simultaneously uncovers how much we resemble them, in our own thinking and behaviour?

  39. Hi Dananjay —
    Yes, it’s interesting how we wish robots to be more like us instead of letting them be their own thing. I’m sure when the times comes, robots will create new robots in their image and not ours. 😀

Comments are closed.