Post Au1P3oeL2fD9Uk14L2 by grammargirl@zirk.us
(DIR) More posts by grammargirl@zirk.us
(DIR) Post #Au1P3oeL2fD9Uk14L2 by grammargirl@zirk.us
2025-05-12T15:07:34Z
1 likes, 2 repeats
This whole article about who can and can't detect AI writing is fascinating! The average person essentially does no better than chance, but a panel of 5 people who use AI for writing every day got only 1 in 300 wrong. The screenshot describes how they look at the writing differently. https://arxiv.org/pdf/2501.15654
(DIR) Post #Au1P3wrOW0PMxYBbnc by grammargirl@zirk.us
2025-05-12T15:12:28Z
1 likes, 0 repeats
Here's an absolutely wild detail! To use paired articles on the same topics, they started with human-written articles and then had AI generate articles of the same topic and length, so it had to make up names for people in the stories, and it used the same names ... a lot:"Annotator 1 is the only one to pick up on 'AI names': in fact, 63.3% of GPT-4O and 70% of CLAUDE-3.5-SONNET articles include either the name Emily or Sarah."
(DIR) Post #Au1PIAo1SjmQGVOMzo by futurebird@sauropods.win
2025-05-12T15:18:29Z
0 likes, 1 repeats
@grammargirl Researchers could get a lot more attention if they made a fun test for people to try— although such a test would quickly be outdated. I don’t feel confident in spotting AI I just know its weaknesses I think a bit more than most people. A little.
(DIR) Post #Au1iKQHCoPQbxGpQDg by MegaMichelle@a2mi.social
2025-05-12T18:51:43Z
0 likes, 0 repeats
@futurebird @grammargirl I want to do a test so bad!
(DIR) Post #Au1jywBHi9JeRUwRQO by roblosricos@social.vivaldi.net
2025-05-12T19:10:14Z
0 likes, 0 repeats
@futurebird @grammargirl i noticed several websites - one which dealt exclusively with science news - which were AI-generated, because the headlines often didn't match the expectations of the content. also, the AI-generated websites were very often quite redundant - presented the same information several times, even going so far as to repeat certain paragraphs verbatim, more than once.