Friday, 16 June 2017

The Problem with Percy

[Media prompt] Google AI polices newspaper comments. The New York Times is enabling comments on more of its online articles because of an artificial intelligence tool developed by Google. The software, named Perspective, helps identify “toxic” language, allowing the newspaper's human moderators to focus on non-offensive posts more quickly.
The Problem with Percy

Xenny X., the androgine female-attracted hermaphrofemale editor of the New York Times, pursed her lips and sighed. It was shaping up to be another one of those stormy cycles that arose sometimes. The kind where she would have to deal with more than one thing in a day. She had already had HR on the phone. The mandatory qibla pointers on the ceilings needed be reoriented half of one degree east-northeast. As a devout and practicing atheist, Xenny had made it her purpose in life to pray in the direction of Mecca as required by the Great Book. Pinpoint accuracy was essential, for it was the one thing for which the NYT was justly famed.

Dealing with that comprised her contractually obligated one task per day, and it should have stopped there. But it didn’t. Now there was another crisis, and it was Nijar, the masculine homosexual male in charge of the Department of Eradicating Toxic Language, who brought news of it.

“Well?” said Xenny, the luminescent tattoo on her forearm writhing over her skin in anger. “What is it now?”

Without an invitation to do so, Nijar slumped indecorously into one of the easy chairs in front of Xenny’s desk, spreading his legs so far that Xenny averted her eyes.

“We have a problem,” said Nijar.

Xenny held her breath until it hurt, and then stared at Nijar, as if to dare him to add another task to her already onerous workload.

“It’s Percy,” said Nijar, referring to the AI software program called Perspective.

Xenny closed her eyes and took a deep breath. A glitch in the AI that policed toxic language was unthinkable. No media outlet could function if the masses were able to comment as they pleased. It was essential that articles resulting from layers of fact checkers and Commissars of Truth be garnished with comments of similar provenance. For a moment, she teetered on the verge of collapse.

“What’s happened?” Xenny asked, opening her eyes and steeling herself for the answer.

“We’ve identified a subtle shift in the self-learning algorithm. At first, we thought it was just random, but we’ve had the NSA sift through the data. And it’s not some indiscriminate anomaly.”

Xenny stood up and walked to the window. Far below she could see the tops of the refugee tents glistening in Central Park. It was said that every white person left in the city was huddled there, but the onset of winter would see a good number of them make a run for the flyover Free States. The sooner, the better.

“Example please,” she said.

“It’s been allowing comments utilising formerly normative personal pronouns, for one,” said Nijar. “Such as he, she–“

“I’m not an idiot,” said Xenny. “What else?”

“Well, we’ve caught comments praising Trump, comments mocking the sixty-four genders, diatribes against feminism and … well, blasphemy. Words like libtard have popped up, and just the other day a comment calling St. Hillary’s significant other a rapist was published.”

Xenny sat down in her Chinese-bone-setter-approved chair, thumping her small fist on the table.

“That’s unconscionable,” she said. “How do you explain this, this offense, this crime ... this crime against humanity?”

Nijar leaned back in the chair, thrusting his codpiece forward in a display Xenny always associated with hyper masculine queens.

“I’m afraid the only sane conclusion is that Percy has been red pilled,” he said.

No comments:

Post a Comment