Across different domains, a growing number of websites are incorporating social features. This study shows that the mere presence of social cues on a website (such as functions for “liking” content or commenting) can cause users who perceive the website as trustworthy to expose themselves to potentially harmful consequences. We carried out an experiment utilizing a YouTube-like video platform that provides the opportunity to study users’ behaviors and perceptions in a realistic, controlled environment. Our results show that, among users who were primed to perceive the website as trustworthy (as opposed to untrustworthy), those who were exposed to social features disclosed more personal information compared with users who were not exposed. Moreover, among high-trust participants, the effect of social features on information disclosure was mediated by participants’ perception that they can connect to other people on the platform. Moreover, the presence of social cues did not influence participants’ privacy concerns.