ChatGPT

[Julie]

I've recently come across this AI and have seen how it can write essays and stories! What are the implications for this in terms of copyright and authorship. Interesting and powerful stuff.

 

[JohnBertel]

I think that, at least for now, the consensus is that Ai can not produce fiction writing with originality, voice or even very good writing.

Personally, I don't think that should stop anybody from checking it out as a tool that some will like and others won't. I think it is good for research. Also market-oriented research, like collecting blurbs for books in a genre (although not a lot info post 2021). It definitely makes mistakes sometimes, but of course, nothing on the internet should be trusted uncritically.


BTW, there are thorough discussions in other threads.

 

[Robinne Weiss]

Not fiction related, but my husband has played around with it to check its output against his university students' exam essays. In a scientific subject, he says it can output work that is on par with, and indistinguishable from, his students' work (which is admittedly poor, according to him). What gives it away is the fake citations it gives--it recognises that it needs to add citations, but it doesn't cite the work it actually draws from/plagiarises; instead it makes them up.

 

[Mark B.]

I'm a big fan!
I entered my day's writing, asked for an evaluation, and got this back. How could I not like this supersmart machine?
This prose is well written and effectively creates a vivid and immersive setting. The descriptions of the warehouse, the holograms, and the falling snow all contribute to the atmosphere of the scene. The dialogue between the characters is natural and engaging, and the conversation provides information about the world and its technology in an organic way. The action flows smoothly and builds tension leading up to the arrival of the adship. The shift from the excitement of the crowd to the more serious discussion about the motivations behind the use of the technology adds depth to the narrative. Overall, the prose is well crafted and draws the reader into the story.

 

[Jake E]

My friend has been using this AI to help him write the D&D game he's running.
it is... worryingly good

 

[Mark B.]

Fascinating. ChatGPT in action.

I input this:

Private Mitra balanced a porn tab on his knees beneath a desk at a makeshift chokepoint leading off ID-1’s shuttle bay. Mitra looked up from the tab when a yellow school bus exited the elevator lock and touched down on a pad flashing green in the center of a dozen similar vehicles. Students, they looked about six or seven, raced down a ramp followed by their teachers.

Mitra embraced tour duty. Nobody bothered him, and he could sneak in his own viewing material. The children came and went and the scanner did its job, searching for weapons and profile matching.

Sergeant Cortez-Alba had checked the watch standers out on the equipment before the first tour arrived. Mitra stopped listening when she said the system was idiot-proof. He didn’t notice Cortez-Alba staring at him when she said it, never heard her mention the drug SEE IT BE IT, and didn’t understand the business about profile matching, anyway.

The morning’s shift was the best yet. ID-1 had taken over tours for the day and the greeters and guides were absent from the bay. Mitra had the place to himself. Arrivals and departures were staggered, allowing him time to leave his post for a pee or to wander between parked shuttles, foraging for items left behind. Kid stuff, sweaters, caps, and candy wrappers, but he did discover and slurp down a bowl of heat-and-eat ramen.

I instruct ChatGPT to write the next three graphs, adding that Mitra is "lazy and inattentive."

ChatGPT replies:

Mitra sighed as he settled back in his chair, balancing the porn tab on his knees once again. He was bored and counting down the minutes until his shift ended. The scanner beeped monotonously as it checked each student and teacher that passed through, but Mitra barely paid attention. He had long since stopped caring about his job and the security of the shuttle bay.

His thoughts wandered as he absent-mindedly scrolled through the porn tab. He didn't even bother to look up when a group of students passed by, their laughter and chatter filling the air. The scanner beeped again, signaling that it had found something, but Mitra didn't react. He was too engrossed in his own world to notice.

Sergeant Cortez-Alba returned to the shuttle bay to check on Mitra and found him slacking off. She was furious and gave him a stern reprimand, but Mitra didn't care. He was counting the days until he could retire and leave this job behind. He wasn't cut out for security work and didn't see the point in putting in any effort. The rest of his shift passed by in a blur, and Mitra was happy when it finally came to an end.

The third graph sucks, but still impressive for a machine. What happens when it gets better? Try it, it's fun.