

I really miss the old wild west of reddit, but you’re totally right it was a much more insane and hostile place for anyone not conforming to the circlejerk. These kinds of smaller communities starting out are also not made for the kinda late adopters that may be wanting to come over and not even just lurk but doomscroll for hours. People will have the best experience if they find a dopamine hit in creating and sharing original content or they can handle lurking for a few minutes at a time and then going out and touching grass the rest of the day.





I only skimmed the article, but I’m familiar with this area of work. It’s the fine tuning phase of llms that require human “graders” in a sense to tell the model what is a helpful vs unhelpful response and elevate it from a literal autocomplete tool to a chatbot. The biggest platform for this kind of work is DataAnnotation if you want to look it up. It’s not just random people in the global south but also professionals training the model on their niche expertise (e.g., lawyers, accountants, engineers). The other type of work that has been going on even pre-llm chatbot era is, as you said, moderation. You moderate the chatbot for “bad” answers, correct them, and often that means impersonating the bot to the end user. I believe this is the main driver behind llm progress lately; literally humans teaching AI how to appear more human.