Ethan Mollick on the study replicating old LLM conclusions
AI summary
Ethan Mollick praised the idea of using new models to replicate old LLM research findings at scale, arguing that much of the existing understanding is outdated: recent models no longer fabricate citations, and both hallucination rates and stereotypes have declined, but sycophantic tendencies and a left-leaning stance still persist.
Body
This is a fantastic idea - a mass attempted replication of old findings of LLMs with new models. Much conventional wisdom is outdated.
Recent models no longer invent citations and both hallucination rates & stereotyping has dropped. But they are still sycophantic & lean left. https://t.co/sPZZDFYAf4