Was the same argument made against dictionaries and the standardisation of spelling?
Standardisation reduced variation, but made writing more interoperable. And LLMs can be remarkably useful for close reading: interrogating claims, checking evidence and exposing gaps in an argument. The claim that AI removes the intellectual work involved in writing seems too broad.
I agree AI is useful in many cases that you mentioned. Here I was focusing on mostly AI generated text, where from relatively small prompt you get a much larger text. In this case it's not only about standardising the text but usually expanding it beyond original intention or not capturing precise enough original idea. So I think it's a bit different to dictionaries, where representation changes but meaning is kept intact. Here for me problem that often meaning is slightly twisted.
The distinction between standardising representation and expanding an underspecified prompt into a much longer text is doing a lot of work here. If the latter is the real concern, it seems important to make that explicit, rather than treating the resulting problems as properties of AI writing in general.
Was the same argument made against dictionaries and the standardisation of spelling?
Standardisation reduced variation, but made writing more interoperable. And LLMs can be remarkably useful for close reading: interrogating claims, checking evidence and exposing gaps in an argument. The claim that AI removes the intellectual work involved in writing seems too broad.
I agree AI is useful in many cases that you mentioned. Here I was focusing on mostly AI generated text, where from relatively small prompt you get a much larger text. In this case it's not only about standardising the text but usually expanding it beyond original intention or not capturing precise enough original idea. So I think it's a bit different to dictionaries, where representation changes but meaning is kept intact. Here for me problem that often meaning is slightly twisted.
The distinction between standardising representation and expanding an underspecified prompt into a much longer text is doing a lot of work here. If the latter is the real concern, it seems important to make that explicit, rather than treating the resulting problems as properties of AI writing in general.