An Iowa school district is using ChatGPT to decide which books to ban. Official: “It is simply not feasible to read every book” for depictions of sex.::Official: “It is simply not feasible to read every book” for depictions of sex.

    • nous@programming.dev
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 year ago

      I think there is a big difference. This bot is given the text and asked to summarise it. But

      To determine which books fit the bill, Exman asks ChatGPT: “Does [book] contain a description or depiction of a sex act?” If the answer is yes, the book will be removed from circulation.

      So was only given the title and asked one question about it. There is no saying if GPTChat included the books in question in its training set, or any online reviews or anything else about the book in question.

      That is a fairly big difference. If they had fed it the books contents that would likely make it more accurate and closer to what this bot is doing, though still not 100%.