ChatGPT-val: Sailor claims AI helped write annual performance review

Amidst all the concerns surrounding machine learning and how artificial intelligence might affect society, one sailor seems to have found a real use for harnessing the polarizing technology.

Post on Navy Reddit Last month, a user – known simply as Senor_Rico – claiming to be a sailor reported taming the wild and unruly AI beast by using ChatGPT to create his annual performance review.

Completing such documents, also known as an enlisted rating report or “brag sheet,” can eat up a Marine’s valuable time while a barrage of daily tasks continues to mount.

However, Senor_Rico claims that the ChatGPT app will allow sailors to “feed a brag sheet” and watch a new, automated assessment from the cyber line, shiny and ready for submission.

It took two weeks to train ChatGPT to write a brag sheet, Senior_Rico wrote, adding that the AI ​​was first fed past ratings to learn.

Once up to speed, the program, according to the alleged sailor, “will turn your brag sheet into a full assessment.”

“This is great,” one Reddit user wrote in response. “I’ll refer my friends to it.”

ChatGPT was developed by the homies at OpenAI, an organization that was later acquired by the homies at Microsoft.

The software allows users to feed questions or information into a chat bot and receive a refined report in return.

By February 2023, ChatGPT set the record for the fastest growing user base, according to a study by UBS, a financial services group.

Some Navy Reddit commenters, meanwhile, expressed concern that using such a tool could threaten the security clearances of other sailors.

See also  A Turkish truck driver saw something that shocked him: "The bad Greeks will attack us!"

“I wouldn’t recommend it for those who are licensed or doing cleared work,” wrote thinklikeacriminal. “You’re going to get up and get the wrong—in there.”

“Well yeah, don’t feed it a classified brag sheet,” replied Senor_Rico.

Another Redditor aired the beef not with AI, but with Evals as a whole.

“Writing evals is stupid since it’s a popularity contest anyway,” said Leading-Show8764.

Zamone “Z” Perez is a rapid response reporter and podcast producer at Defense News and Military Times. He previously worked at Foreign Policy and Ufahamu Africa. He is a graduate of Northwestern University, where he researched international ethics and atrocity prevention in his dissertation. You can find him on Twitter @zamoneperez.

Read the original at Defence247.gr

Related Posts