A new study finds that humans and AI spot different kinds of deepfakes — hinting at the need to team up to fight them.
As submissions to AI conferences soar, authors’ own rankings — combined with peer review — could offer a powerful way to identify high-impact research.