•   over 1 year ago

GPTZero is Not Reliable

In the email today it was mentioned the organizers checked for AI generated answers using GPTZero. It's well known among academia now that it is a highly unreliable tool with lots of false positives (and negatives).

Here are some such examples in the news:

https://www.usatoday.com/story/news/education/2023/04/12/how-ai-detection-tool-spawned-false-cheating-case-uc-davis/11600777002/

https://www.reddit.com/r/Professors/comments/10pya6w/gpt_zero_false_flagging_human_written_papers/

https://futurism.com/gptzero-accuracy

Not looking for changes in the results: it is clear what the level of organization this hackathon has. Hopefully this information would be useful to others in the future.

Signing out, flesh and blood, zero AI used,
RowHammer.

  • 1 comment

  • Manager   •   over 1 year ago

    Hi Yogeshwar Bala,

    Understand your concerns.
    We had our Experts to review the responses first and have their comments.
    Later the GPTZERO AI score was taken as a reference for further filtering.

    Regards,
    Team MAH !

Comments are closed.