It’s not what you say, but how you say it: Using artifact intelligence to detect deception in natural speech in an HR recruiting context
研究利用自然语言处理技术，仅通过分析语音判断说话者是否在说谎，准确率约82%，可辅助HR在招聘中识别虚假陈述，且设计注重数据隐私。
Human emotions trigger physical reactions of the body that can be interpreted using artificial intelligence (AI) methods. AI-enabled detection mechanisms offer new opportunities to gain deeper understanding of human emotions. AI is already able to recognize basic emotions, such as joy, anger, or fear, but is challenged to detect complex emotions accurately, such as when someone is lying. The human voice conveys a wealth of subconscious information. This research utilizes natural language processing (NLP) to present a novel AI artifact that analyzes solely the human voice to detect whether a person is speaking to their true conviction. This is useful for use cases where it is important for decision makers to know whether a person’s arguments are truthful. A suitable use case could be corporate recruitment, an area where people traditionally lie a lot. The artifact can support human resource (HR) personnel in making better decisions by overcoming their naturally weak ability to detect deception. To suit the sensitive context of recruitment talks, the artifact is designed to protect data privacy, and thus has no speech recognition capabilities and does not store information that identifies the speaker. The model detects deceptive statements with ∼82% accuracy. Integrating the artifact into corporate business processes would enable managers to more accurately detect deceptive behavior of their interview partners. The increasing use of novel AI artifacts to compensate for humans innately weak capabilities to detect other people’s complex emotions will challenge our theoretical conceptualization of organizational decision making in light of emerging human–AI hybrids.