Ethics, Bias and Natural Language Processing

Prof. Dr. Kevin Tang, Summer 2025, Course Catalog

Course Description

Is technology really as innocent and as objective as they are said to be? As machine learning (ML) and Artificial Intelligence (AI) becomes more prominent in our life from speech and voice recognition by Alexa to automatic fake news warnings of social media posts, issues with social bias and fairness in language technology become more pertinent than ever before. Negative impacts that biased ML and AI could have for various social identities such as race, gender and culture.

Through research papers, we will gain a better understanding of the the ethics, fairness, bias-related challenges in Natural Language Processing. Throughout the course, students will gain an overview of the various types of Natural Language Processing models (e.g., Automatic Speech Recognition, Language Model, hate speech detection) and its implications for bias, diversity, Inclusion, Environmental and human costs, privacy and governance. They will gain comprehensive insights into the following key areas:

  1. Dangers and superpowers: e.g., ”On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?” (Bender et al. 2021) https://dl.acm.org/doi/10.1145/3442188.3445922

  2. Bias: e.g., ”Language (Technology) is Power: A Critical Survey of “Bias” in NLP” (Blodgett et al. 2020) https://aclanthology.org/2020.acl-main.485/

  3. Diversity and Inclusion + Environmental and human costs: ”A Survey of Race, Racism, and Anti-Racism in NLP” (Field et al. 2021) https://aclanthology.org/2021.acl-long.149/

  4. Privacy + Governance and Participatory Approaches, e.g., ”Data Governance in the Age of Large-Scale Data-Driven Language Technology” (Jernite et al. 2022) https://dl.acm.org/doi/abs/10.1145/3531146.3534637

In the course, we will sample from existing research papers. You will be asked to read, discuss and summarize your understanding of the papers, through i) short reflection posts (paragraph-size), ii) selected longer research summaries (one page) per theme, and iii) in-class discussions .

The in-class discussions will be held in different formats, such as i) poster presentations, ii) group debates, iii) role-playing, iv) mind-map.