

Yeah this, but also people are seriously not a monolith based on made up demographic categories. Some Latinos identify as white and/or american only and are racist/xenophobic against other brown immigrants. Some Latinos are from communist countries so the red scare works on them. Some Latinos identify so strongly with being “good” people that when Trump talks about getting rid of the illegals and criminals, they take those words at face value. There’s so many really fucking dumb naive people out there that vote, Latinos included.

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.