“There may be some glitches, as the simulation for the neural patterns for linguistic processing is still new and unstable.”
• • • •
“David?”
All the faces of you—the phases of you. I will never be tired of them; have enough of them all every entire. The lingering light of a September afternoon; the smell of popcorn and hotdogs. Nervous. Will you or won’t you? The promise of the premise. Then I see you. And there is no more holdback suspense doubt. A softness that curls into me, fits me in all the right places. Complete. Warm. Sweet. I will yes I will yes.
“Dad!”
Little fingers, delicate, ramified tendrils reaching extending stretching reaching into the dark ocean that you once drifted in; a smile the heat of a thousand suns.
I cannot conceive you. A missing presence, a wound in the mouth of the heart that the tongue of the will cannot stop probing. I have always have missed missed missed you, my darling.
“What
happened
to him?”
“He died. You were there, Ellen. You were there.”
“Then
what
is this?”
“I suppose you’d call this an example of unintended consequences.”
“You’d better start making some sense.”
More text scrolled across the screen:
Integrating placement and routing; NP-complete; three-dimensional layout; heuristics; fit and performance; a grid, layers, the flow of electrons in a maze.
“Logorhythms supplies the world’s best chips for high-volume data processing. In our work, we often face a class of problems where the potential solution space is so vast, so complicated, that it’s impractical for even our fastest computers to find the best solution.”
“NP-complete problems,” said Maddie.
Dr. Waxman looked at her.
“Dad explained them to me.”
That’s my girl.
“Right. They show up in all kinds of applications: circuit layout, sequence alignment in bioinformatics, set partitioning, and so on. The thing is, while computers have trouble with them, some humans can come up with very good solutions—though not necessarily the best solution—very quickly. And David was one of them. He had a gift for circuit layout that our automated algorithms could not touch. That was why he was considered our most important resource.”
“Are you talking about intuition?” her mother asked.
“Sort of. When we say ‘intuition,’ often what we mean are heuristics, patterns, rules-of-thumb that can’t be articulated because they’re not consciously understood as such. Computers are very fast and very precise; humans are fuzzy and slow. But humans have the ability to extract
insight
from data, to detect patterns that are useful. It’s something that we’ve had trouble recreating with pure artificial intelligence.”
Maddie felt a chill in the pit of her stomach.
“What does this have to do with my dad?”
Faster, faster. Everything is so slow.
Dr. Waxman avoided looking at her. “I’m getting to that. But I have to explain the background to you—”
“I think you’re just dragging this out because you’re ashamed of what you’ve done.”
Dr. Waxman stopped.
My girl.
Dr. Waxman gave a light chuckle, but there was no mirth in his eyes. “She’s impatient, like you.”
“Then get to the point,” Mom said. Dr. Waxman started at the icy intensity in her voice. Maddie reached for her mother’s hand. Her mother squeezed back, hard.
Dr. Waxman took a deep breath, blew it out. “All right,” he said in a resigned monotone. “David was ill; that was true. You remember that he died during surgery, the final attempt to save him that you were told had very little chance of success?”
Mom and Maddie nodded together. “You said only the clinic at Logorhythms could do it because it was so advanced,” Mom said. “We had to sign those liability waivers for you to operate.”
“What we didn’t tell you was that the surgery wasn’t intended to save David’s life. His condition had deteriorated to the point that the world’s best doctors couldn’t have saved him. The surgery was a deep scan of his brain, meant to save something else.”
“A
deep scan
? What does that mean?”
“You’ve probably heard that one of Logorhythms’s moonshot projects is to completely scan and encode the neural patterns of a human brain and to recreate them in software. It’s what the Singularity nuts call ‘consciousness uploading.’ We’ve never succeeded—”
“
Tell me
what
happened to my husband!
”
Dr. Waxman looked miserable.
“The scan, because it needs to record neural activity with such detail… requires destruction of the tissue.”
“
You cut up his brain?
” Mom lunged at Dr. Waxman, who held up his hands in a vain attempt to defend himself. But the screen had come alive again, and so she stopped.
There was no pain. No no no pain. But the undiscovered country, oh, the undiscovered country country.
“He was dying,” said Dr. Waxman. “We were absolutely certain of that before I made the decision. If there was a chance to preserve something of David’s insights, his intuition, his skill, however slim, we wanted—”
“You wanted to keep your top engineer as an algorithm,” said Maddie, “like a brain in a jar. So that Dad would go on working for you, making money for you, even after he died.”
Die, die, die. DIE.
Hate.
Dr. Waxman said nothing, but he lowered his face and hid it in the palms of his hands.
“Afterward, we were very careful. We tried to re-encode and simulate only the patterns we believed had to do with circuit layout and design—our lawyers wrote us a memo assuring us that we had the right since the know-how was really Logorhythms’s intellectual property, and didn’t belong personally to David—”
Mom almost lunged out of her chair again, but Maddie held her back. Dr. Waxman flinched.
“Did David make a lot of money for you?” She spat the words out.
“For a while, yes, it appeared that we had succeeded. The artificial intelligence, which modeled the extracted portions of David’s technical know-how and skills, functioned as a meta-heuristic that guided our automated systems very efficiently. In some ways, it was even better than having David around. The algorithm, hosted on our data centers, was faster than David could ever hope to be, and it never got tired.”
“But you didn’t just simulate Dad’s intuition for circuit layout,
did you?
”
The wedding dress; layers of lines. A kiss; a connection. The nightstand, the Laundromat, the breath on a winter’s morning, Maddie’s red apple delicious cheeks in the wind, two smiles in a flash—a thousand things make up a life, as intricate as the flow of data between transistors nanometers apart.
“No.” Dr. Waxman looked up. “At first, it was just odd quirks, strange mistakes the algorithm made that we thought were due to errors in identifying the parts of David’s mind that were relevant. So we loaded more and more of the rest of his mental patterns into the machine.”