Artificial intelligence fairness in the context of accessibility research on intelligent systems for people who are deaf or hard of hearing

Kafle, Sushant; Glasser, Abraham; Al-khazraji, Sedeeq; Berke, Larwan; Seita, Matthew; Huenerfauth, Matt

doi:10.1145/3386296.3386300

Cited by 11 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, when conducting empirical work, researchers should consider type, severity, and the co‐occurrence of multiple disabilities when developing participant sampling and recruitment plans. Moreover, when seeking to understand applicant reactions to, and experiences with, technology, it may also be important to consider the intersectionality between disability status and other individual characteristics given the known impact of AI on many diversity groups (Bogen, 2019; Kafle et al, 2020). A partnered research approach (Bonaccio et al, 2018; Fisher et al, 2022), involving a variety of perspectives including people with lived experience of disabilities, can help researchers develop more interesting research questions, create more rigorous and inclusive research designs, and improve participant recruitment.…”

Section: Discussionmentioning

confidence: 99%

“…Similarly, applicants who use adaptive devices or functions such as captioning may be at a disadvantage in interacting with selection technologies. AI‐generated captions are not currently as reliable as professional live captioning (Kafle et al, 2020). These authors note that the AI‐based systems fail in unpredictable ways, and it is difficult for users to determine if the information provided is accurate.…”

Section: Theorizing About Applicants With Disabilities’ Reactions To ...mentioning

confidence: 99%

See 1 more Smart Citation

Reactions of applicants with disabilities to technology‐enabled recruitment and selection: A research agenda

Fisher,

Connelly,

Bonaccio

2023

Int J Selection Assessment

View full text Add to dashboard Cite

Technology‐enabled recruitment and selection technologies, such as chatbots, assessment games, and asynchronous video interviews, are becoming more widely used. However, their impact on people with disabilities is frequently ignored; this has potentially significant implications for the perceived fairness of hiring decisions. We advance eight theoretical propositions on the positive and negative implications of technology‐enabled recruitment and selection technologies for applicant reactions of people with disabilities. Our propositions are based on three key design features of these technologies: separation in time and space, automated administration, and automated evaluation. We provide recommendations for future research and discuss practical implications for the use of advanced technology in recruitment and selection.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Theorizing About Applicants With Disabilities’ Reactions To ...mentioning

confidence: 99%

Reactions of applicants with disabilities to technology‐enabled recruitment and selection: A research agenda

Fisher,

Connelly,

Bonaccio

2023

Int J Selection Assessment

View full text Add to dashboard Cite

show abstract

“…AI-based interaction techniques have enabled various new forms of human-machine interfaces such as chatbot, audio speech recognition, and gesture recognition, which might result in accessibility problems for disabled people. Mitigation of such unfairness has aroused concerns in recent research [202,368].…”

Section: Human-ai Interactionmentioning

confidence: 99%

Trustworthy AI: From Principles to Practices

Li¹,

Pan²,

Liu³

et al. 2021

Preprint

View full text Add to dashboard Cite

Fast developing artificial intelligence (AI) technology has enabled various applied systems deployed in the real world, impacting people's everyday lives. However, many current AI systems were found vulnerable to imperceptible attacks, biased against underrepresented groups, lacking in user privacy protection, etc., which not only degrades user experience but erodes the society's trust in all AI systems. In this review, we strive to provide AI practitioners a comprehensive guide towards building trustworthy AI systems. We first introduce the theoretical framework of important aspects of AI trustworthiness, including robustness, generalization, explainability, transparency, reproducibility, fairness, privacy preservation, alignment with human values, and accountability. We then survey leading approaches in these aspects in the industry. To unify the current fragmented approaches towards trustworthy AI, we propose a systematic approach that considers the entire lifecycle of AI systems, ranging from data acquisition to model development, to development and deployment, finally to continuous monitoring and governance. In this framework, we offer concrete action items to practitioners and societal stakeholders (e.g., researchers and regulators) to improve AI trustworthiness. Finally, we identify key opportunities and challenges in the future development of trustworthy AI systems, where we identify the need for paradigm shift towards comprehensive trustworthy AI systems.

show abstract

“…As far as we know, there have been only limited work to tackle the issue of fairness in speech technologies. Koenecke et al [10] identified significant performance gaps between white and African American speakers, while [11] and [12] pointed the necessity to include data from people with disabilities in current training sets.…”

Section: Related Workmentioning

confidence: 99%

ASR4REAL: An extended benchmark for speech models

Rivière¹,

Copet²,

Synnaeve³

2021

Preprint

View full text Add to dashboard Cite

Popular ASR benchmarks such as Librispeech and Switchboard are limited in the diversity of settings and speakers they represent. We introduce a set of benchmarks matching reallife conditions, aimed at spotting possible biases and weaknesses in models. We have found out that even though recent models do not seem to exhibit a gender bias, they usually show important performance discrepancies by accent, and even more important ones depending on the socio-economic status of the speakers. Finally, all tested models show a strong performance drop when tested on conversational speech, and in this precise context even a language model trained on a dataset as big as Common Crawl does not seem to have significant positive effect which reiterates the importance of developing conversational language models.

show abstract

Artificial intelligence fairness in the context of accessibility research on intelligent systems for people who are deaf or hard of hearing

Cited by 11 publications

References 13 publications

Reactions of applicants with disabilities to technology‐enabled recruitment and selection: A research agenda

Reactions of applicants with disabilities to technology‐enabled recruitment and selection: A research agenda

Trustworthy AI: From Principles to Practices

ASR4REAL: An extended benchmark for speech models

Contact Info

Product

Resources

About