With ubiquitous exposure of AI systems today, we believe AI development requires crucial considerations to be deemed trustworthy. While the potential of AI systems is bountiful, though, is still unknown-as are their risks. In this work, we offer a brief, high-level overview of societal impacts of AI systems. To do so, we highlight the requirement of multi-disciplinary governance and convergence throughout its lifecycle via critical systemic examinations (e.g., energy consumption), and later discuss induced effects on the environment (i.e., carbon footprint) and its users (i.e., social development). In particular, we consider these impacts from a multi-disciplinary perspective: computer science, sociology, environmental science, and so on to discuss its inter-connected societal risks and inability to simultaneously satisfy aspects of well-being. Therefore, we accentuate the necessity of holistically addressing pressing concerns of AI systems from a socioethical impact assessment perspective to explicate its harmful societal effects to truly enable humanity-centered Trustworthy AI.
Since the fatal shooting of 17-year old Black teenager Trayvon Martin in February 2012 by a White neighborhood watchman, George Zimmerman in Sanford, Florida, there has been a significant increase in digital activism addressing police-brutality related and racially-motivated incidents in the United States. In this work, we administer an innovative study of digital activism by exploiting social media as an authoritative tool to examine and analyze the linguistic cues and thematic relationships in these three mediums. We conduct a multi-level text analysis on 36,984,559 tweets to investigate users' behaviors to examine the language used and understand the impact of digital activism on social media within each social movement on a sentence-level, word-level, and topic-level. Our results show that excessive use of racially-related or prejudicial hashtags were used by the counter protests which portray potential discriminatory tendencies. Consequently, our findings highlight that social activism done by Black Lives Matter activists does not diverge from the social issues and topics involving police-brutality related and racially-motivated killings of Black individuals due to the shape of its topical graph that topics and conversations encircling the largest component directly relate to the topic of Black Lives Matter. Finally, we see that both Blue Lives Matter and All Lives Matter movements depict a different directive, as the topics of Blue Lives Matter or All Lives Matter do not reside in the center. These findings suggest that topics and conversations within each social movement are skewed, random or possessed racially-related undertones, and thus, deviating from the prominent social injustice issues.
Recently there are increasing concerns about the fairness of Artificial Intelligence (AI) in real-world applications such as computer vision and recommendations. For example, recognition algorithms in computer vision are unfair to black people such as poorly detecting their faces and inappropriately identifying them as "gorillas". As one crucial application of AI, dialogue systems have been extensively applied in our society. They are usually built with real human conversational data; thus they could inherit some fairness issues which are held in the real world. However, the fairness of dialogue systems has not been investigated. In this paper, we perform the initial study about the fairness issues in dialogue systems. In particular, we construct the first dataset and propose quantitative measures to understand fairness in dialogue models. Our studies demonstrate that popular dialogue models show significant prejudice towards different genders and races. We will release the dataset and the measurement code later to foster the fairness research in dialogue systems.