AIHackadayabout 1 hour ago
Language Tips To Avoid Anthropomorphising LLMs
2 min read
Emily M. Bender and Nanna Inie suggest language changes to avoid anthropomorphising LLMs, such as "automatic transcription" instead of "speech recognition" and "input"/"output" over "prompt"/"response". They admit "undesirable output" for "hallucination" is not entirely appropriate.
Level
Hype check
Tap to vote and see what everyone thinks.
Summary by ByteBrief
