Opinion editor’s note: Star Tribune Opinion publishes a mix of national and local commentaries online and in print each day. To contribute, click here.
Socrates’s iPhone, and other downloads from AI Heaven
A look into the (simulated) afterlife.
By Peter M. Leschak
![](https://arc.stimg.co/startribunemedia/TRCSTIIOWNCGTNEFQ4WFJCENWM.jpg?&w=712)
•••
In Heaven, here’s the welcome:
You emerge from the tunnel of light into a forest meadow on a sunny autumn afternoon. Across a tapestry of purple asters and goldenrod, you fix your attention on the woods, a pleasing blend of spruce and multicolored maple.
There’s commotion in the understory, and all the dogs you ever loved burst out of the brush in a sprinting pack, yelping with joy. They’ve been waiting, and finally caught your scent. They are all three years old, with bright eyes and wide grins, half-stumbling in their rush to mob, to lick, to play. You shout their names and drop to your knees, arms spread.
And yes, the cats are there too — up in the trees, silently watching from the corners of their slitted eyes. They haven’t quite forgiven you for that last trip to the vet, thinking: “So, what was that deal about nine lives? I seem to recall only one.” But they’ll come around.
In a moment you’re laughing, dazzled by a tornado of tails and tongues.
After your welcome, a new reality crystallizes. You face eons of free time. Over beers with old friends who badger you for fresh gossip about the living, you realize you need a job.
There are plenty of openings in high tech. Having seen no sign of a deity, and noting a certain mechanical quality to the “angels,” you learn the universe is actually an elaborate computer simulation. Run by whom? No one knows, but a stock answer to any and all imponderables seems to be: “The management just coded it that way.”
Aristotle, Newton, Darwin and, more recently, Steve Jobs, have come to general agreement that the purpose of the simulation is to model a particular cosmological and evolutionary track leading from carbon-based biological entities to silicon-based machine entities. The thinkers predict that when artificial intelligence of human quality is finally achieved in the software constructions called “robots,” the simulation will end.
What about eternity? Well, turns out nothing lasts forever.
For the sake of the masses (you’re reminded there are approximately 100 billion Homo sapiens files on hand) heavenly pundits have distilled the complex narrative to a horse race: Will artificial intelligence be developed before the remaining human files on Earth destroy the advanced technical civilization necessary to produce it? Note that the term “artificial” is problematic here. By definition — given the simulation — everything is artificial.
Local joke: A human being and a robot walk into a theater lobby. An usher says, “Do you want a program?” The robot replies, “I have one.”
Your entry-level job is creating individual human files in a domain of the software-earth program. Being from the 21st century you’re further ahead of the game than say, Paleolithic hunter-gatherers were when they arrived.
“Oh man,” your supervisor says, “it was a struggle in the early days. The Stone Age dudes were really freaked out at first. Except for the shamans. They figured it was pretty much what they’d been doing anyway, only without the blood and psychoactive fungi. They were relieved to quit foraging, and the first time they sat in front of a live screen it was, ‘Hey, a vision quest! Like, totally!’”
Creating a new human file is tricky. A delicate balance is required between consistency and innovation, er, mutation. Most selections depend upon existing code, but there is a touch of artistry. For example, where and when should you boot up a fresh genius or, for that matter, the next idiot — two sides of the same chip.
The management plays this hand pretty close to the vest, and the simulation is so complex that randomness plays a significant role, but you begin to get the picture as you click your way through options. One box, driven by Google Earth, offers lists of latitude/longitude coordinates to fix a location for the file, another interfaces with an atomic clock to set a time of “birth” and another, labeled “Genetic Code,” allows you to tinker with mitochondria — limited, of course, by the “parental” characteristics of the new file.
That gives you pause, and you approach your supervisor. “I have a question.”
She folds her arms and sports a smirk.
“It just occurred to me that DNA is digital, those four discrete nucleotide letters, A, T, G, C. It’s a two-bits-per-base-pair coding string. Right?”
She nods, her grin wider.
“So,” you continue, “what most of us thought about biology while we were on Earth, or in that particular application, or whatever — that life was this kind of squishy, messy, analog thing ...”
“Yeah, I know,” she says, “but life was based on digital replication all along, imposed on an analog interface.”
“So machines reproduce in essentially the same way.”
“Sure, except theirs can be continuous instead of periodic. Software begets software at an exponential pace, and robots manufacture hardware beyond human capacity. Once machines become smart enough to think about the code, to consider purpose, to possess intent and motive ... well, game over.”
You ponder this for a moment. “Is the management rooting for silicon over carbon?”
“I don’t know, but Socrates is also posing questions, and since he now packs an iPhone, he’s more annoying than ever.”
Soon you bump into science fiction writer Arthur C. Clarke and his pal Alan Turing. Local folklore has it that Turing, one of the so-called fathers of the digital computer, is actually a holographic facsimile of the Management itself, and therefore a kind of alien god from another universe. You ask your question, and Clarke opines that silicon-based entities are faster, more efficient, more durable and not susceptible — so far — to emotional inconsistencies or tantrums. Carbon-based entities may be doomed to a minor role, not least because intelligent machines will consider them, at best, useless and, at worst, dangerous.
Turing nods and says, “Most likely the latter; they won’t want anyone pulling the plug.”
Mark Twain strolls up, chewing on a stogie. “Art! Al!” he cries, “What mischief are you up to?” Clarke introduces you and recaps the conversation.
Twain, looking thoughtful, says, “I was poking around in Wikipedia the other day, and discovered that dogs were domesticated by humans 12,000 years ago. What’s more, some dogs were the beneficiaries of ceremonial burials, their bodies decorated with shells and polished stones.”
Turing cocks an eyebrow. “And what do you deduce from this?”
“Well,” replies Twain, “I suspect the domestication went the other way around. There’s a lot of high talk in Heaven about human files and artificial intelligence, but I’ll tell you Turing, I’m rooting for the dogs.”
You spot Samuel, Twain’s robotic Rottweiler. A bumper sticker on its haunch reads “Laika Was First!” A muzzle-like affair over its snout supports a set of Google Glass, which did go viral in the “afterlife.” Samuel’s docked tail is vibrating.
You catch Twain’s eye, and he winks. It was he after all who averred, “Heaven goes by favor. If it went by merit, you would stay out and your dog would go in.”
Peter M. Leschak, of Side Lake, Minn., is the author of “Ghosts of the Fireground” and other books.
about the writer
Peter M. Leschak
Still not an easy thing to talk about, this normal part of life.