CAREER: Emotion Artificial Intelligence in the Future of Work: A Privacy and Relational Ethics Lens

情感人工智能(AI)的目标是在平凡和高风险的背景下推断人们的情绪和情绪。尽管存在有效性、偏见和隐私问题，但情感人工智能在没有知情的公众意见的情况下，继续以相应的方式开发和使用，例如在就业面试中。对于它可能对人们的隐私和生活造成的好处和危害，我们缺乏经验知识。这个项目调查了情感人工智能在两个高风险的背景下--工作场所和招聘面试--的伦理和隐私影响。该项目将描述情感人工智能是如何使用的，以及人们可能会如何受到它的影响。该项目具有政策含义，旨在解决这项技术对工人和求职者的影响，从而影响工作的未来。此外，这项研究为情绪人工智能的创建者和决策者提供了指导方针，并提高了公众对情绪人工智能的使用和含义的了解。本项目调查了情绪人工智能是如何被求职者和员工感知和体验的，以及应用它产生的推理的决策者是如何使用情绪人工智能的。首先，该项目采用了混合方法，包括使用故事完成法和因素情节调查的访谈。这些研究的参与者包括被边缘化的员工和更容易受到情绪人工智能伤害的求职者。其次，该项目包括采访和一项与决策者的实验，以检查他们可能如何使用情绪人工智能。这项研究将对情感人工智能/人工智能伦理、隐私、工作的未来和情感计算学术做出贡献。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

Emotion artificial intelligence (AI) aims to infer people's emotions and mood in mundane and high-stakes contexts. Despite validity, bias, and privacy concerns, emotion AI continues to be developed and used in consequential ways, such as in employment interviews, without informed public input. We lack empirical knowledge about the benefits and harms it can inflict on people's privacy and lives. This project investigates emotion AI’s ethical and privacy implications in two high-stakes contexts--the workplace and hiring interviews. The project will describe how emotion AI is used and how people may be affected by it. The project has policy implications addressing this technology’s impact on workers and job seekers, and therefore the future of work. Additionally, this research leads to guidelines to be used by emotion AI creators and decision-makers, and improves public knowledge of emotion AI’s use and implications.This project investigates how emotion AI is perceived and experienced by job seekers and employees and how it is used by decision-makers who apply the inferences it generates. First, the project uses a mixed-methods approach including interviews using the story completion method and factorial vignette surveys. Participants in these studies include marginalized employees and job seekers at more risk of emotion AI harm. Second, the project includes interviews and an experiment with decision-makers to examine how emotion AI may be used by them. This research will contribute to emotion AI/AI ethics, privacy, future of work, and affective computing scholarship.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.