I imagine they're expanding a bunch of notes and reports and research into an article, whose total size is not that much bigger than the "meat" of the inputs. The annoying LLM fluff is largely when someone is trying to inflate some simple information in size by an order of magnitude, e.g. turning three bullet points into an article. If you have enough of actual content to furnish an article, well, LLMs are excellent at restyling that into coherent prose.
alt Hacker News
I imagine they're expanding a bunch of notes and reports and research into an article, whose total size is not that much bigger than the "meat" of the inputs. The annoying LLM fluff is largely when someone is trying to inflate some simple information in size by an order of magnitude, e.g. turning three bullet points into an article. If you have enough of actual content to furnish an article, well, LLMs are excellent at restyling that into coherent prose.