How do people watch AI-generated videos of physical scenes?

TL;DR

This study investigates how viewers' gaze behavior is influenced by the perceived authenticity of highly realistic AI-generated videos of physical scenes, revealing that awareness of AI origin shifts viewing from passive to active anomaly detection.

Contribution

It provides the first analysis of human gaze behavior in response to AI-generated physical scene videos, highlighting the impact of perceived authenticity on viewing strategies.

Findings

Gaze behavior is influenced more by perceived authenticity than actual video authenticity.

Awareness of AI generation prompts viewers to actively search for anomalies.

High realism in AI videos affects viewer engagement and trust.

Abstract

The growing prevalence of realistic AI-generated videos on media platforms increasingly blurs the line between fact and fiction, eroding public trust. Understanding how people watch AI-generated videos offers a human-centered perspective for improving AI detection and guiding advancements in video generation. However, existing studies have not investigated human gaze behavior in response to AI-generated videos of physical scenes. Here, we collect and analyze the eye movements from 40 participants during video understanding and AI detection tasks involving a mix of real-world and AI-generated videos. We find that given the high realism of AI-generated videos, gaze behavior is driven less by the video's actual authenticity and more by the viewer's perception of its authenticity. Our results demonstrate that the mere awareness of potential AI generation may alter media consumption from passive viewing into an active search for anomalies.