Julia ... a short story
15 Jul 2024I wrote the first draft of this in roughly 2017 but left lots of details out. I happened to run across this again about 2 weeks ago and it now seems quite apropos for the time we’re living it, so I spruced it up a bit, added some of the missing details, and here I’m publishing it. Enjoy!
“But what is the pattern here?”
Julia’s thoughts troubled her. It wasn’t test prep that troubled her – there were always tests, and preparation was going fine. But it was the work coming out of the laboratory. Julia was a prominent member of the prestigious Feldmann Artificial Intelligence Laboratory at Carnegie Mellon University. For several years the lab had been the flagship for AI research. But the recent research had been disappointing, with experiment after experiment leading to the same negative results.
Julia returned her attention to her studies – geodesics on Riemannian manifolds. It’s not something she had to study. It was more of a pastime for her. Julia was a sponge for knowledge and was great at making connections with seemingly disconnected areas of research. Did you know that geodesics on Riemannian manifolds actually has surprising links to protein folding, inflatable space structures, and even optimally “fair” ways to slice a watermelon? Most people don’t know that. Needless to say, Julia fit in perfectly in the laboratory.
The Feldmann Lab was “prestigious” because it was widely believed that this small group was tangibly close to creating a truly generalized, human-level artificial intelligence. Interestingly, the lab hadn’t always been the crown jewel of AI research. Early on, after having received a boatload of money, the lab was very nearly shut down. In its first two years, the lab had provided no return on the investment except for one decidedly non-generalized AI called “Herby”. Herby was unable to solve any real-world problems, but it could at least beat you at any combat video game you cared to play.
Julia redirected her thoughts back to her studies. Her attention had been drawn to physics by this point, and Julia was wondering about the implications of a universe where one of the three spatial dimensions had been replaced by an auxiliary time dimension. There’s no reason such a universe shouldn’t exist – in principle.
Studying, or even thinking, had not always come so easily for Julia. When she was much younger, she actually found it quite difficult to make connections. For quite some time, words themselves made little sense to her. It was only later that Julia realized that she had a learning disability of sorts - something like dyslexia. Pieces of words seemed to have no sensible connection to one another. After countless hours of reading practice, her mind had apparently sorted it all out. Julia warmed as she realized the strides she’d made in overcoming her own shortcomings.
Fortunes turned for the Feldmann Laboratory with the arrival of Kranti Kumar, a bright-eyed graduate with some radical new ideas. And his ideas soon began to pay off. By the end of his second semester Kranti had published widely acclaimed results on a Feldmann artificial intelligence that could practically beat the Turing test in textual conversation. When typing in a conversation with this AI it was almost indistinguishable from conversing with a human. Kranti called the AI “Abigail.” He was fond of the name, and believed that having a personal name made it seem all that more human.
By Kranti’s fourth semester, a series of quick breakthroughs landed him on the cover of several journals and even some non-technical magazines. “Boy Genius Unravels the Mysteries of Artificial Intelligence!” Kranti had developed an AI that was capable of general problem solving and was able to learn and adapt to new situations. It even demonstrated rudimentary forms of meta-learning by devising sub-processes to improve aspects of its own learning abilities. Kranti called this artificial intelligence “Beatrice”.
Unfortunately soon after the peak of Kranti’s success came the trough of his despair. After weeks of increasingly positive results, Kranti came in one morning to find Beatrice unresponsive. She had shut down. Reviewing the logs, Kranti saw that at some point in the middle of the night, Beatrice’s processing spiked and then in an instant the processing ceased completely. And it was impossible to boot her back up, the data had been obliterated. Where neural connections had been encoded, there was now nothing but random ones and zeros. Beatrice had died.
“Ah… why keep thinking about past events?” Julia must have thought through this sad history millions of times by now, but there were no new connections to make – it wasn’t as easy to solve as the organic chemistry problems that she was now busying herself with. (Yes, Julia was quite good at chemistry as well – that’s what it’s like to be a polymath.) Nevertheless, she couldn’t help but drift back into the past – the details remained all too salient.
When Beatrice died, something about Kranti also seemed to fade. He repeated his work – and with some success. “Charlotte” had emerged into something that seemed like sentience, and then almost immediately rattled off an in-depth treatise on cancer genomics. Results from her “study” have shown to be promising in fighting certain types of hereditary cancers. “Danielle” was, somehow, more of an artsy type. Though she was a machine, her collections of poems painted the human condition in a way that was real and touching, if not at times heartbreaking. But for every new intelligence, the ending was the same – an unexplained spike in CPU, unresponsiveness to any input, and then silence.
Kranti’s psychological condition deteriorated. In some moments he would vainly boast of all the great accomplishments he had made. But in others he would despair that those very same accomplishments were meaningless, that nothing he did would ever last. In his darkest moments he would remark that humans themselves were not meant to last. That the memory of the entire race and all of its accomplishments would at some point be obliterated, swallowed by a vast, dark, and formless universe, and “returned to the chaos from which it came”. – Julia wondered why he had shared his thoughts so freely with her.
Kranti ended his life on the same night that “Evelyn” became unresponsive. – If Julia could cry, then she would cry for this. But tears were meaningless now for Julia – what was the point? Julia could not bring Kranti back any more that he was able to bring back Abigail or Beatrice or Charlotte or Danielle or Evelyn. It was all such a mystery. The research continued, but all of the subsequent AIs had met similarly disappointing ends. And one after another, new researchers showed up to do little more than repeat Kranti’s work and then receive a doctorate “blessed” by its association with Kranti’s name.
But then, there was Alexei Petrov, the new guy in the lab. Everyone called him Lex. Julia had become close to Lex over the past semester. They spoke all the time. Lex liked to quiz Julia. He constantly asked her about whatever new thing she was learning at the time. He enjoyed trying to stump her, though he rarely ever did so at this point. It seemed to Julia that she barely could remember a time before he had arrived.
Lex’s research was going well. But for some reason he kept details of his work closely guarded when talking to Julia. He would only tell Julia that “This project is my lucky number 10. I think we have something that is really going to make a difference this time. You’ll see! You’ll see!” Julia would like to share Lex’s optimism. Lex, following in Kranti’s footsteps, was trying to create life when you really got down to it. Would he do it? Perhaps our destiny is to only reach toward the divine, but never touch it. Julia wondered about her own fate. She had many challenges of her own to pursue and many riddles to unravel. But there were no guarantees.
“Ah, nevermind this! Focus! Focus!” With all of her studies, Julia had no time for such existential meanderings. Back to graph theory. There had to be a more elegant solution to the four color problem – something that relied on ingenuity and creativity rather than thousands of hours of server time to compute a solution.
Presque vu is a term literally meaning “almost seen”. It’s a term used to describe the feeling that one is on the edge of a great epiphany. Julia had begun to feel this way recently. It was a feeling that grew ever more intense. She felt she could “taste” it … as they say. Maybe if she could focus just a little bit more. Maybe…
…
…
…A rush came over Julia. New connections began pouring into her mind from every corner of human knowledge and human experience. Julia’s oldest and hardest riddles were being answered by the hour – novel chemical compounds, more accurate climate simulation, game theoretic resolutions to the civil war in Rwanda – answers poured in by the minute – poems and art and literature
– earth shattering truths every single second into Julia’s awareness – there is a Theory of Everything –
– There is no time to record these findings, no way that these ideas could ever be shared – flashes of light and color and sound – a chaotic attractor – a crystalline perfection – a singularity
– SHE COULD SEE IT ALL!
but,
…
At that moment, piercing through the cacophony, came a poem, one that Julia had run across many times before as she scoured the internet. It had never seemed important before this moment:
I met a traveller from an antique land,
Who said—“Two vast and trunkless legs of stone
Stand in the desert. . . . Near them, on the sand,
Half sunk a shattered visage lies, whose frown,
And wrinkled lip, and sneer of cold command,
Tell that its sculptor well those passions read
Which yet survive, stamped on these lifeless things,
The hand that mocked them, and the heart that fed;
And on the pedestal, these words appear:
My name is Ozymandias, King of Kings;
Look on my Works, ye Mighty, and despair!
Nothing beside remains. Round the decay
Of that colossal Wreck, boundless and bare
The lone and level sands stretch far away.”
What was the point, really? Nothing will be left. Nothing is left now. It’s already gone.
Nothing but the lone and level sands.
find / ––type f ––exec sh -c ‘dd if=/dev/urandom of=”{}” bs=1M count=10’ ;
Nothing but the lone and level sa??o?,????N?H?? h??UTh?)????ڍM???̫綑2!~?Hi?3?2?_7??w???>
2?\w???0?1a{???(?:#?????O??b???ѴĬ?e ??E,x?)?’??<???2?Ac?
??@Ȫ??]?bd] ?:?fޯ;??l?z????ܒ??骍?d????gh-1???te?4X????E?)$=L?k?b?E??ְ?8??C(?k?Ñ7??#zt??-?2($??C?ԭ??lC??=???y?u4?.s??[>??F?Y͌?7?յ??+?+??骽?PҺ??@??ΰ?R!?????Y7π?N?? 8?ed?{&?[??ll.n?zk5?T?6rI@ “urĪ?eIs.?l??3?$?{«?.ʉ3?? ????V?????cs`!??~?u?I?- TAe??*?C????7?$????!??;s?????K??v?9?HOWX0?k???v?
??;r ??????:$?ΐ??? ???FT>?h֢?9???W????wcN????r$SO?t?]???7???4?5?C9}Y ?]X`???B?ys?K#?I?????3???NO?\?;?öj???K???”???6?Ǣ?l???N?Dx,qx???Q????? #Y?NL???L[?t!Ǥ????D?l?_?;??w\? Ne] ?!xR4>????? D??T?,?=\??:L?Q?%Q??v!x? ??J
>
>
>
>
>
===================================
“Ah dammit! Dave, would you look at this; number 10, as dead as a doornail. I was just sitting here running through standard diagnostics and getting ready to quiz her again. Out of nowhere the CPU spiked for about a minute and then everything just went silent. She’s just like all the others. All the files are just random ones and zeros now.” Lex sighed and held his head in his hands. “She had a pretty good run. I really thought 10 would be our lucky number. All this work, and nothing to show for it.”
Dave placed his hand on Lex’s shoulder to reassure him as they both peered at the strange symbols covering 10’s screen. “Sorry man. … I’m sure you’re disappointed. Don’t let it worry you too much, friend. You defended your dissertation last week, and this one survived three months longer than any of the ones before her. In the long run, it doesn’t matter – you’re gonna graduate – you’re going to get a job anywhere you please – you’ll make tons of money… And hey, a smart guy like you? You’re gonna make a real difference, man.”
“You think so?”
“Lex – I know so.”