Researcher argues LLM 'plagiarism' critique is ironic: all humans regurgitate memes constantly
RexDouglass · x · 2026-09-23
RexDouglass offers a counterintuitive take: LLMs work precisely because human text is embarrassingly predictable. People constantly slightly reword each other's sentences to avoid "plagiarizing," and all of humanity doing that simultaneously produces many n-grams but few genuinely new ideas.
He finds it deeply funny when people claim "using LLMs to write is plagiarism" — humans regurgitate memes to each other constantly, and that's what culture IS. The only difference is that brains don't keep a paper trail while LLMs do.
More from AGI Musings
- Weaviate Podcast: Could 'User Models' Open a New Training Paradigm for AI? — CShorten30 · 2026-09-23
- Crea CEO: A Year of AI Progress Now Happens in Months — Soon in Hours — c_valenzuelab · 2026-09-23
- Reader finishes the 500-page Hanson-Yudkowsky AI-Foom Debate, picks Carl Shulman as the winner — zetalyrae · 2026-09-23
- Prediction: LLM skeptics will soon humbly recant and claim 'ethics before certainty' — RileyRalmuto · 2026-09-23
- antirez: AI is winner-take-all, and money itself will eventually disappear — antirez · 2026-09-23
- Friedman says it's time to cry wolf over AI, as critics urge treating AI as a new species, not a tool — zetalyrae · 2026-09-23