Social Robot Interactions for Social Engineering: Opportunities and Open Issues

摘要

The field of information security is receiving greater attention in recent years. It is well-known that Social Engineering attacks exploit psychological manipulation techniques to trick people in providing private information without being aware. Robots are entwined in our daily lives and humans tend to trust them more and more. This leads to a kind of "overtrust" in robots that in turn may expose humans to security, fraud and privacy issues. This paper aims to provide insights into how interaction with Social Robots could be exploited for the purpose of Social Engineering. In particular, the paper presents the ability of robots to gather information during an interaction/conversation with humans and how this information could be integrated and enriched with Emotion Recognition techniques. Authors show some preliminary experimental results that are encouraging since the interaction between the robot and the users resembles a humanto-human conversation. In particular, for Emotion Recognition has been obtained two different results: an accuracy of 74.38% fusing two biometric traits (voice and face) and a tolerance of 5 bpm to estimate heart-rate.