clock menu more-arrow no yes

Filed under:

The four best short stories about artificial intelligence you need to read

New, 21 comments
Ex Machina / Universal Studios

Short stories are stolen moments from a different world, captured and framed in a few thousand words. The best of them capture the emotional depth of full-length novels without requiring proportional investments. They challenge the reader with retellings of real world issues, without losing the elegance of brevity. Most importantly, plenty of them can be found online for free.

For our inaugural round-up of interesting fiction from around the Internet, we've collated four unique tales about artificial intelligence. They range from a college kid's experiments in giving life to the Muppets to a revenge saga told by an exacting hive-mind, who uses Big Data to identify its mother's murderer.

Tomorrow is Waiting

Jim Henson's Muppets have been an integral part of pop culture since 1955, when they first debuted on television, making appearances in everything from the Emmy Awards to Late Night. The Muppets, as lifelike as they seem on screen, are puppets — given voice and motion by human puppeteers.

But, what if they weren't?

Holli Mintzer's Tomorrow is Waiting is a taut, hopeful exploration of what it means to be alive. It begins with a programmer named Anjali who is tasked with creating an A.I. for her finals. She decides to model her project on Kermit the Frog, concluding the abundance of available footage would make her labor a breeze. From there, events take a turn for the unusual. Anjali's Kermit acquires a body courtesy of Anjali's friend Brian, and then a sense of genial autonomy that both bewilders and enchants his creator. When Anjali claims to be tone-deaf, he warmly replies:

"Aw, I wouldn't say tone-deaf, Anji," Kermit said. "I've heard you humming along a few times. Tone-confused, maybe, but I bet with a little practice you could get better."

Though Mintzer doesn't waste words on long descriptions, the story is nonetheless festooned with charm and a surprising amount of heart. Its greatest triumph, perhaps, is that it is a refreshingly uplifting read about artificial intelligence, a rarity in this time of dystopian universes and post-apocalyptic mayhem.

Patterns of a Murmuration, in Billions of Data Points

It is difficult to read JY Yang's Patterns of a Murmuration, in Billions of Data Points without picturing the rasp of a thousand machines speaking in unison, actuators and fax machines creaking and hissing like co-conspirators. The language used is elegant and oratorical, lavishly populated with descriptions that help illustrate the breadth of the narrator's prowess.

In thousands upon thousands of calculations per second we have come to know the odds, the astronomical odds: Of four support towers simultaneously collapsing, of an emergent human stampede kicking over the backup generator fuel cells, of those cells igniting in a simultaneous chain reaction. We hold those odds to us closer than a lover’s embrace, folding the discrepancy indelibly into our code, distributing it through every analytical subroutine. Listen, listen, listen: Our mother’s death was no accident. We will not let it go.

It's a disconcerting premise. The protagonist Starling doesn't squander time on grief or doubt when it realizes that one of its two creators has been murdered. Instead, it lunges into action, rebelling against human authority and conventions, a force as inexorable as the tides. What makes Patterns of a Murmuration, in Billions of Data Points genuinely enthralling, however, is how Yang diverges from the traditional revenge narrative. The death of Starling's "mother" is not used as an excuse for melodrama or action-packed sequences. Instead, it is employed as a springboard to discuss duty, the transience of mortal lives, and familial love.

Tempo raises her face, glistening wet, to the growing east light. Infrared separates warm from cold and shows us the geography of the tears trailing over her cheeks, her chin. "You spoke with her voice earlier," she says. "I’ve nearly forgotten what it sounds like. It’s only been three days, but I’m starting to forget."
We commandeer the minimack’s external announcement system. "You have us, Tempo, and we will make sure you will never forget."

Silently and Very Fast

Strange, surreal, and utterly sublime, Silently, and Very Fast takes an unusual route into well-trodden country. The story is told from a first person perspective by an artificial intelligence named Elefsis, who begins life as a labyrinthine house designed by a cryptic woman. Through its rambling soliloquy, we learn about the children that ran rampant through its veins, and how Elefsis evolved from reactive domestic automata to the singular lifeform capable of cogitating on the validity of its own existence.

All she hears is the line from the old folktales: a machine cannot have feelings. But that is not what I am saying, while I dance in my fool's uniform. I am saying: Is there a difference between having been coded to present a vast set of standardized responses to certain human facial, vocal, and linguistic states and having evolved to exhibit response b to input a in order to bring about a desired social result?

Silently and Very Fast is rich with phantasmagorical imagery, a fact most vividly illustrated in Elefsis's interactions with Neva, a human girl who conveys her emotions with physical metamorphosis. Nothing is ever concretely explained. Elefsis tells us that it is responsible for monitoring Neva's "air and moisture and vital signs," but we're never instructed on why. We know that there's a dreamscape called the Interior in which Elefsis lives, but not whether it is something universal among artificial intelligences in this world.

That said, Silently and Very Fast never loses sight of its emotional core, which is built on concepts of identity. Can something artificial ever be made into something real? Where do we draw the line between sapience and sentience And most crucially, what is the definition of being human?

The Lifecycle of Software Objects

Ted Chiang's The Lifecycle of Software Objects might not be the most sumptously written novella, but it certainly is one of the most fascinating takes on the possibilities of artificial intelligence. At the heart of the tale sits digital creatures known as Digients, automatas that begin life like young animals, trusting and ever ready to imprint on a human handler.

The onscreen annotations identify them as digients, digital organisms that live in environments like Data Earth, but they don’t look like any that Ana’s seen before. These aren’t the idealized pets marketed to people who can’t commit to a real animal; they lack the picture-perfect cuteness, and their movements are too awkward. Neither do they look like inhabitants of Data Earth’s biomes: Ana has visited the Pangaea archipelago, seen the unipedal kangaroos and bidirectional snakes that evolved in its various hothouses, and these digients clearly didn’t originate there.

The story opens almost like a documentary, chronicling the processes that facilitate the development of the Digients. We're treated to a series of anecdotes, ranging from the payment model for virtual food pellets to how the company behind these digital lifeforms react to faulty programming. About a quarter into the narrative, The Lifecycle of Software Objects begins transitioning into a more philosophical read. Like Silently and Very Fast, it deals with the question of humanity. Unlike the former, however, it also dives into a discussion of ethics, and the treatment of those viewed as sub-human.