Humans cannot reliably distinguish LLM-generated news from human-written news across multiple models, with domain expertise providing only modest help and fatigue reducing accuracy over time.
Miles McCain, and Miles Brundage
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CY 1years
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Can Humans Tell? A Dual-Axis Study of Human Perception of LLM-Generated News
Humans cannot reliably distinguish LLM-generated news from human-written news across multiple models, with domain expertise providing only modest help and fatigue reducing accuracy over time.