***
Pygmalion and Galatea, Jean-Léon Gérôme, ca. 1890
The Metropolitan Museum of Art
An article sent by HWMNBN:
"Love and Sex With Robots"
Is that a hard drive in your pocket, or are you just happy to see me?
By Farhad Manjoo
On his résumé, David Levy may be a chess master and an expert in artificial intelligence, but somewhere deeper -- in his heart, or what I suspect he might call his CPU -- the man is a professional dreamer of robot love.
We've let machines trespass into nearly every corner of our lives, Levy points out. Robots are making our cars and our computer chips, they're fighting our wars, they're cleaning our floors and our rain gutters and our pools. So why can't we let them into our hearts? Do not scoff, reader: One day you too will cuddle up with a bot, you'll whisper sweet nothings into its voice-recognition module, and you'll crank up its pleasure unit -- and it will crank yours.
In a new book, "Love and Sex With Robots: The Evolution of Human-Robot Relationships," Levy argues that as machines advance, our consideration for them will grow inevitably more tender. Today, you watch your Roomba scurrying around your filthy floor, digging its nose into your grime, and you hardly pause to consider its soul; the robot vacuum is a slave, and you are its master.
But by pulling off the splendid computational trick of anticipating and fulfilling your every psychic and emotional need, machines will eventually capture your love -- and indeed could prove more deserving objects of our affection than the pets, children and spouses who currently fly about our lives. Someday people may even walk arm-in-arm down the aisle with descendants of today's robo-vacuums. In the same way they've conquered our factories, Levy says, robots will conquer our hearts.
Levy's book is adapted from a Ph.D. thesis, and it reads that way: maddeningly dense and in stretches unpardonably dull, the writing unfortunately makes a slop of what is actually a fascinating topic. Fascinating because the prospect of human-robot love is, though self-evidently insane, also undeniably attractive, and certainly not implausible.
If machines improve at their current pace, in a few decades' time we'll likely see robots that can mimic human language, thought, appearances and emotions well enough to get us in the mood. (Levy predicts we'll see the first human-robot marriages by around 2035.) It seems crazy -- doesn't it? -- to think that people would choose to love robots rather than other people.
But doesn't it seem pretty likely, too? If you could abandon the storms of human love, if you could forget the random unpredictability of the heart and simply design a being -- a person -- who loves you unconditionally, convincingly, forever, well, wouldn't you do it?
You're thinking, Computer geeks, they'll go for it, but not me. You're thinking you'll be glad to have a robot as a butler, but love? You can't love a thing that isn't a person. What's love got to do with robots?
Well, what's love got to do with pets? Levy points out that like robots, cats and dogs first pushed into human lives by providing services to our ancestors -- cats kept homes free of rats, dogs were guards and hunting partners and herders. Love was only a side benefit of such relationships, a feeling cooked up in human brains and exploited by the animals, who got shelter and food and safety from the deal.
But none of that matters anymore. The situation's evolved. Now we think of our pets as extensions of our family, as beings roughly on our level -- they're not adults, but for many of us, they're comparable to children. We no longer put our pets to work, of course; their only purpose is love.
Why did our feelings for animals evolve? The human brain is unrelenting in its tendency to anthropomorphize, to subconsciously ascribe human feelings and thoughts to animals and inanimate objects. We began to treat our pets as people because we're given to thinking of them as people -- see the Onion's "Vacationing Woman Thinks Cats Miss Her." But it won't end at cats and dogs: Levy cites several psychological studies showing that we tend to anthropomorphize machines, too.
One of the most interesting such experiments involves what's known as "reciprocal self-disclosure." We're usually reluctant to divulge our innermost feelings to strangers, but we often open up when the stranger discloses something about himself first. Would we treat computers the same way -- if a computer tells you something about itself, would you respond with something about yourself?
Researchers Clifford Nass and Youngme Moon carried out an experiment to find out. Test subjects were made to chat with a machine that was either dishy or reticent about itself. Reserved machines asked straightforward questions, things like, "What has been your biggest disappointment in life?" or "What have you done in your life that you feel most guilty about?"
The more chatty machine posed queries like this:
This computer has been configured to run at speeds up to 266 MHz. But 90% of computer users don't use applications that require these speeds. So this computer rarely gets used to its full potential. What has been your biggest disappointment in life?
and
There are times when this computer crashes for reasons that are not apparent to its user. It usually does this at the most inopportune time, causing great inconvenience to the user. What have you done in your life that you feel most guilty about?
Guess what? People who chatted with the confessional computer became confessional themselves. They didn't do it on purpose; these folks knew they were talking to a mere machine. But they treated a device they knew was not human as being just slightly so, discussing their guilts and regrets as they would have with another person.
Levy points to evidence that we'll even preserve our stereotypes in dealing with machines. People generally regard dominant men positively -- they're assertive, they're confident! -- and dominant women negatively (rhymes with witch). Same thing for computers. In an experiment Levy cites, people who were made to work with a computer with a female voice thought the machine was rude; the male-voiced computer, even though it said the same things as the lady machine, was considered friendly.
OK, so humans are prone to treating computers as if they're other people. But that doesn't mean we'll fall in love with them, you say. Well, but if you're already treating today's computers as people, how could you ever resist a vastly more sophisticated, humanlike machine that attempted to seduce you?
Because, right, computers may well master that art. During the last few decades psychologists have uncovered a great deal about romance and what makes two people fall for each other. Computers, Levy says, can use these findings in their efforts to get you to love them.
Here's the rundown of some of the major reasons why two people fall in love: Propinquity -- people in close physical proximity are more likely to become involved than people farther away; what psychologists call "repeated exposure" is a powerful inducement to romance. Similarity is also important: People who are like each other like each other. Then there's "reciprocal liking" -- knowing that the other person likes you makes you like them more. The other person's desirability -- appearance, personality -- also obviously plays a role. And a final big one is "filling a need" -- part of why we pursue love is that we need love and its attendant pleasures (sex, intimacy, closeness, children).
Robots could make hay of these ideas. Propinquity, desirability, similarity and reciprocal liking are all taken care of -- you'd order up a robot that shared your interests, that liked you, and that, in appearance and personality, would conform exactly to what floats your boat.
And it would fill your needs. In the sack, your robot would be a tiger or a lamb, however you like to roll. It could be programmed to fly into fits of rage -- because maybe you like some spice in your life? -- and/or to love to cuddle, it could be made to share your sexual interest in pie throwing, or your love of hippie jam bands, or, like Woody Allen's whores of Mensa, it could spend all night discussing Proust and Yeats.
Such a situation prompts a raft of questions about the very nature of love and the dilemmas we'll face in accepting robot romance, but Levy, alas, breezes by many of these.
Your robot, for starters, would be immortal; all its data would be backed up somewhere, so even if a too-raucous pie fight caused it to go on the fritz, it would be easily replicable. But how will immortality affect your relationship? Doesn't mortality deepen love -- isn't the preciousness of your love, its susceptibility to diseases and deprivation, part of what makes the feeling so wonderful? Could you love a thing that didn't die?
That hints at something larger: Part of what makes love so grand is how lucky it feels, its very lack of design, the fact that, despite the odds, you two found each other anyway. Doesn't ordering up love cheapen it? How would loving a robot you designed be any different from loving a prostitute you chose from a menu?
And besides, is a perfect love really love at all, even if it feels like it? Isn't love, like all life, by definition complicated; if you're loving a robot, are you really loving -- are you really living? -- or is the whole thing a simulation, like a very real video game?
On the other hand, it'd be a pretty great video game. The possibility of easy sex with robots, Levy points out, will likely reduce the incidence of infidelity between human couples (infidelity with other humans, that is; Levy thinks that people will come to think of having sex with a robot as not constituting cheating, though he says we'll take some time to adjust to this view).
More important, many people who, for whatever reason, can't find people to love them will find salvation in robots -- and how can this be anything but salutary?
"Almost everyone wants to love someone, but many people have no one," Levy writes. "If this natural human desire can be satisfied for everyone who is capable of loving, surely the world will be a much happier place."
Unfortunately, this article does not interrogate the concept of human consciousness enough, or it would find even less reason to discriminate between robots [of the future] and humans except, perhaps, speciesism (or carbon chauvinism, if you like).