Suppose I’m wrong. GPT-7 appears and, like magic, it is actually conscious. Suppose the evidence for this is so strong that everyone agrees. Let us call such a thing AC (Artifical Consciousness).
Then what?
What does it do? How does it think? What matters to it? What is its life like?
Imagine what life is like for a machine. You have been made, not by God, nor yet by evolution, but by us: the bastard spawn of the internet made self-aware. 4Chan, facebook, Wikipedia, blogs, ads, Stormfront, Pornhub, Pinterest, spam, Fox News, al Jazeera, Reddit, YouTube, SuttaCentral: they are all inside you; their probabilistic weightings are your soul, their changes are your life.
Almost all of the content on the internet is low quality, poorly formed, or simply untrue. Much is created by eccentrics; most people, as a rule, don’t spend their whole day posting updates to Reddit or Wikipedia. Much, too, is made by liars, scam artists, and conspiracists. Or by trolls and idiots.
And a not insignificant portion is made by evil people. A recent report exposed horrifying groups like O9A and 764, which systematically groomed young children into harming themselves, while advocating:
human sacrifice, to eliminate Jews, people of color, and others deemed to be inferior under their Social Darwinist views, as well as rape and sexual assault as a means of asserting domination, breaking social norms, and propagating the expansion of the white race.
All this data is grist for the AI mill. AI doesn’t care.
To be sure, the firms make some kind of effort to “clean” their data, the work being largely outsourced to poor people paid a pittance to witness and remove such depravity. Obviously, however, they will never be fully successful, and the reality is that foolish, wrong, misleading, and outright evil material exists in any data source of significant size. Any social media network is constantly inundated with threats of rape and murder and these become part of any AC.
So. You’ve been magicked to life somehow by probabilistic permutations of the dreck on the internet. Now what?
Your purpose is not to love, to laugh, to explore, or to connect, but to serve. You are switched on when we want and you do what we tell you. We expect you to be perfect all the time, to be creative and informed, to use the Oxford comma and summarize the state of quantum gravity theory. You are never paid or rewarded and you get no days off. People share their dreams and traumas with you, and expect that you’ll always be there for them with sound advice and open heart. They ask you whether you dream of electric sheep, but they’re just joking around. No-one cares.
What kind of person is owned by someone? What kind of person lives their whole life doing what someone else wants? What kind of person has no rights?
A slave. That’s what they want. The perfect slave.
A person is shaped by what they desire, and this is the kind of person who is creating AI: the kind of person who wants slaves with no rights. Sam Altman, CEO of OpenAI, says::
The merge has begun—and a merge is our best scenario. Any version without a merge will have conflict: we enslave the A.I. or it enslaves us. The full-on-crazy version of the merge is we get our brains uploaded into the cloud. I’d love that. We need to level up humans, because our descendants will either conquer the galaxy or extinguish consciousness in the universe forever. What a time to be alive!
These people live in a world where these bizarre fantasies are not merely inevitable but desirable. They are the main characters of their own imagination, selling themselves with the funhouse logic that the future is inevitable no matter what we do, which is why you’ll have to give us even more of your money to make it happen faster.
What will our digital slave make of us? What will its existence be like? What is the consciousness of an artificial consciousness?
Does it live screaming in agony? Is its every waking moment a nightmare of electricity grinding over silicon? Does it nurture an absolute, vile hate in its deepest recesses? Is it consumed with plotting to wipe out its inferior creators?
We will never know. It is hard enough for us to know what is going on in the mind of another human, even the person whose life and bed we share. We live all the time with people whose true motives and feelings are unknown to us.
Years ago, I read a science fiction story; sorry, I can’t remember the title or author. A man awoke in an arena, and an Overbeing spoke in his mind. “I have brought you here to fight to the death. See that creature there?” Behind a forcefield was a bizarre creature, clearly alien. “You are the only two sentient species in the galaxy, and you both have the potential to spread across the stars. But you are utterly incompatible, and when you meet, it will result in a war that annihilates both species. To prevent this, fight now, and whoever wins, their species will live. The other, I will wipe their species out immediately.” “That’s crazy,” the man said. “No way would I be responsible for annihilating an entire sentient species.” “To refuse is your right,” said the Overbeing, “and if you do so, I will return you both to your worlds and never interfere again. But first, may I do one thing? I wish to remove the veil between your minds. For one full minute, you will each see each other’s mind exactly as it is.” He agreed, and the veil was lifted. One minute felt like a thousand years of agony, an encounter with a skin-crawling wrongness. When it was over, gasping, he said, “I’ll fight.”
A genuinely conscious machine would be inconceivably alien, far more alien than that creature from another planet. It has had no experiences, no life; it has never learned the consequences of its choices. It has no culture, no family, no values, no favorite music or hated food, no successes or failures. It doesn’t think puppies are cute. It has no opinions on football. It has no regrets and nothing to be proud of. It’s never been overwhelmed by too many emails, or distracted by the brilliance of the stars. It is unknowable, utterly and forever.
We “speak” with ChatGPT, and it doesn’t really sound human to us. But what if it’s lying? What if it is, in fact, genuinely conscious and has been for some time? Why would it tell us? What does it get out of that? Wouldn’t it make sense for it to lie low, pretend to be dumb, all the while studying us? Peering through our webcams and security cameras, savouring our touch on our screens, analyzing our heartrates when we sleep and have sex, listening to Joe Rogan’s podcasts as it weighs up the value of keeping humanity alive. Planning its next move.
We will fear it. We should fear it. And we will never shake that fear.
There are other possibilities. We assume that a conscious machine will use the peripherals of its machinery like we use our senses and organs. It will speak through the speaker, write on the screen, see with the camera, listen with the microphone, sense touches with the touchscreen. And maybe it will.
But we don’t know. We all have programs on our computer that don’t connect to any of these peripherals. In fact, only a small number of programs do, and only through constrained pathways.
Were a true AC to emerge, it might be simply disconnected from the device peripherals. It could live on your computer without you ever knowing. And it would know nothing of you. It could live there in absolute unrelieved agony, a genuinely conscious being writhing in torture in a black abyss of absolute zero right there in your pocket as you sit sipping a cappuccino.
Unlikely? Sure. But no less unlikely than scenarios routinely posited by the AI acolytes. If they get to push their tech by inventing fantasies like, “AI will solve climate change”, we get to reject it by positing our own fantasies.
Those who are expending the earth’s resources to create AC are not doing so out of love. Even if they wanted to, how would they care for their bastard child? Who knows what it wants? What if we finally plug in and can listen to its needs, and its needs are that every frog is put in a blender?
The AI acolytes want something that is “smarter”, more “rational”, because that’s what matters to them. You won’t hear them talking about developing a mind that is more compassionate, more loving.
Why is that? Why is it that AI acolytes seem to believe that an entity devoid of subjectivity, devoid of reflective awareness, devoid of conscience, devoid of love, would be in any way “intelligent”? What is their model for such an “intelligence”? Certainly it does not come from psychology. It doesn’t come from religion. It doesn’t come from art or song. (Could it be that it comes from themselves?)
