# Words of Wisdom: Representational Harms in Learning From AI Communication

arXiv: https://arxiv.org/abs/2111.08581 · HTML: https://mlsanity.com/paper/2111.08581/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 41 references that are in the map.
- First posted on arXiv 16 Nov 2021.

## Builds on (most cited first, 6 of 6)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [Mitigating Bias in Algorithmic Hiring: Evaluating Claims and Practices](https://mlsanity.com/paper/1906.09208/) (2019)
- [Generating Natural Questions About an Image](https://mlsanity.com/paper/1603.06059/) (2016)
- [Ask to Learn: A Study on Curiosity-driven Question Generation](https://mlsanity.com/paper/1911.03350/) (2019)
- [Pedagogical Agents for Fostering Question-Asking Skills in Children](https://mlsanity.com/paper/2004.03472/) (2020)

## Built on by (most cited first, 4 of 4)

- [Is a Seat at the Table Enough? Engaging Teachers and Students in Dataset Specification for ML in Education](https://mlsanity.com/paper/2311.05792/) (2023)
- [What Is the Point of Equality in Machine Learning Fairness? Beyond Equality of Opportunity](https://mlsanity.com/paper/2506.16782/) (2025)
- [GenQ: Automated Question Generation to Support Caregivers While Reading Stories with Children](https://mlsanity.com/paper/2305.16809/) (2023)
- [A Peek Behind the Curtain: Using Step-Around Prompt Engineering to Identify Bias and Misinformation in GenAI Models](https://mlsanity.com/paper/2503.15205/) (2025)

## Abstract (arXiv)

Many educational technologies use artificial intelligence (AI) that presents generated or produced language to the learner. We contend that all language, including all AI communication, encodes information about the identity of the human or humans who contributed to crafting the language. With AI communication, however, the user may index identity information that does not match the source. This can lead to representational harms if language associated with one cultural group is presented as "standard" or "neutral", if the language advantages one group over another, or if the language reinforces negative stereotypes. In this work, we discuss a case study using a Visual Question Generation (VQG) task involving gathering crowdsourced data from targeted demographic groups. Generated questions will be presented to human evaluators to understand how they index the identity behind the language, whether and how they perceive any representational harms, and how they would ideally address any such harms caused by AI communication. We reflect on the educational applications of this work as well as the implications for equality, diversity, and inclusion (EDI).