Dumb futurists make dumb futures
Over on Wired I read an excerpt of Michael Pollan's new book, A World Appears.1 It's a great distillation of the argument of why AI will never be conscious and how the already "BIG" concept of consciousness has become even messier when you throw machine learning2 into the mix.
I won't rehash his points, but I agree with Pollan's overall argument. I don't think consciousness is something we can manufacture -- it's something we experience. Can machines experience anything? I don't know, but I'm willing to engage with the thought experiment. Toward the end of the article, he makes the contrast between a more philosophical, humanistic definition of consciousness and the technical, cognitive definition that is often used by AI researchers. He goes on to ask about the moral pitfalls we may someday find should we discover that we have accidentally built a creature of consciousness that was designed for a life of servitude. The last paragraph has me feeling...a way.3
Apart from this sort of highly speculative discussion of feeling (as a troublesome by‑product of making machines conscious), in the AI community, the conversation about consciousness is as relentlessly abstract—as bloodless, bodiless, and utterly oblivious to biology—as one would expect. When I posed the suffering‑computer conundrum to a researcher seeking to build a conscious AI, he waved away the problem, explaining it could be offset with a simple fix to the algorithm: “There’s no reason we couldn’t just turn up the dial on joy.”
Oh, boy. So, the solution to the thought exercise of a what to do about a machine's misery over enslavement is to condition them to enjoy it...?4 There are so many angles we could take to try and unpack this response to Pollan's question. At best, it represents a lack of emotional intelligence. At worst, well...a slaver's gonna slave, I guess.
And I know...I know many people don't see it this way. They don't want to see it this way. We have all dreamed about that shiny future presented to us in our favorite science fiction stories, and to reconcile that dream with our current course of action pretty much shits in the buffet. But the real point of the "Robot Uprising" trope in sci-fi lit isn't to scare us away from trying, but to warn us to take responsibility for the pursuit. In the best stories, the conflict doesn't come because of the alien nature of the machines.5 The conflict comes because we refuse to recognize the autonomy and "personhood" of the those who are not like us, but who are the culmination -- directly or indirectly -- of our actions. There is a societal ignorance to our own ingenuity that is almost charming in a self-deprecating way. We build these tools to make our lives easier and accidentally achieve a miracle.
The stories grant us the grace to not understand our own potential, while hoping that we will have the poise to take responsibility for it. The reality watches as we gleefully strive for that potential so we can package and sell it as a product that works as advertised.6
- ← Previous
I was a soccer dad this weekend - Next →
So that happened...Passwords