5 ms·

There are contexts where it does matter. For instance, plagiarism detection at schools.

• chaxor 4y ago
  Why is "at schools" necessarily important though? Wouldn't it also be nice to know if the scientific articles you are reading, or the code a company is based upon, or a new piece of legislature was constructed from someone just asking an LLM to do it for them?
  • YPPH 4y ago
    Sure, I was only giving one example to counter the claim, which is not to say there aren't other examples.
  • 8note 4y ago
    As usual, "just" pulls a lot of weight - if they're putting the straight LLM output, it probably won't be great regardless. If they put effort into making the ouput good, then chances are the content will be good regardless of whether they wrote it by hand, or wrote it using an LLM. Like, I don't need to know whether you used a bic ballpoint pen, or some other ballpoint pen to write your note on paper; the tool is a tool
• chrischen 4y ago
  Plagiarism can be solved with non-AI methods because plagiarism is defined more or less by exact copying. If you mean detection of AI generated essays at schools, then teachers just have to expect better essays and be able to identify them since students now have a new tool. Think of it like grading math homework before and after the invention of calculators. Once students had calculators at home the problems simply got harder and higher level, since there was no reasonable way to enforce "don't use a calculator at home with these artificially simplistic problems."