3 ms·
This entire article is the bad toupee fallacy. Readers think they don't like LLM-authored text because they only recognize bad LLM-authored text as LLM-authore
by Bratmon 28d ago
This entire article is the bad toupee fallacy.
Readers think they don't like LLM-authored text because they only recognize bad LLM-authored text as LLM-authored.
Blind trials have actually shown that readers generally prefer LLM authored books to human-authored ones on the same subject.
- cbg0 27d agoThe "trials" you're referring to is probably this study on 1000-word short stories https://www.cambridge.org/core/journals/judgment-and-decision-making/article/bot-or-not-can-people-tell-the-difference-between-stories-written-by-a-human-or-by-an-ai-system/45E6DC0BB90AA648654D5AE243F6C667 https://www.cambridge.org/core/journals/judgment-and-decisio...