About

Biased technology disadvantages certain groups of society, e.g. based on their race or gender. Recently, biased machine learning has received increased attention. Here, we address a different type of bias which is not learnt from data, but encoded during the design process. We illustrate this problem on the example of Conversational Assistants, such as Amazon's Alexa, Apple's Siri, Microsoft's Cortana, or Google's Assistant, which are predominately modelled as young, submissive women. According to UNESCO, this bears the risk of reinforcing gender stereotypes.

In this project, we will explore this claim via psychological studies on how conversational gendering (expressed through voice, content and style) influences human behaviour in both online and offline interactions. Based on the insights gained, we will establish a principled framework for designing and developing alternative conversational personas which are less likely to perpetuate bias. A persona can be viewed as a composite of elements of identity (background facts or user profile), language behaviour, and interaction style. This framework will include state-of-the-art data-efficient NLP deep learning tools for generating dialogue responses which are consistent with a given persona. The persona parameters can be specified by non-expert users in order to to facilitate more inclusive design, as well as to enable a wider critical discussion.

People

This project is a collaboration between researchers at Heriot-Watt University, the University of Edinburgh, and Strathclyde University.

Heriot-Watt University

Verena Rieser

Matthew Aylett

Gavin Abercrombie

Tanvi Dinkar

University of Edinburgh

Judy Robertson

Valentina Andries

University of Strathclyde

Benedict Jones

Victor Shiramizu

Research output

Teaching materials

• Smart Speakers and AI. 2023

  • University of Edinburgh and Newbattle High School. Video.

  • Alan Wait, Valentina Andries and Judy Robertson. Teacher's Guide.

Publications

• Sachin Sasidharan Nair, Tanvi Dinkar, Gavin Abercrombie. 2024. "Exploring Reproducibility of Human-Labelled Data for Code-Mixed Sentiment Analysis". ReproHum 2024. Association for Computational Linguistics.

• Tanvi Dinkar, Gavin Abercrombie, Verena Rieser. 2024. "ReproHum #0927-03: DExpert Evaluation? Reproducing HumanJudgements of the Fluency of Generated Text". ReproHum 2024. Association for Computational Linguistics.

• Valentina Andries and Judy Robertson. 2023. Alexa doesn't have that many feelings: Children's understanding of AI through interactions with smart speakers in their homes. Computers and Education. 

• Gavin Abercrombie, Amanda Curry, Tanvi Dinkar, Verena Rieser, and Zeerak Talat. 2023. Mirages. On Anthropomorphism in Dialogue Systems. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 4776–4790, Singapore. Association for Computational Linguistics.

• Gavin Abercrombie, Aiqi Jiang, Poppy Gerrard-Abbott, Ioannis Konstas, and Verena Rieser. 2023. Resources for Automated Identification of Online Gender-Based Violence: A Systematic Review. In Proceedings of the 7th Workshop on Online Harms and Abuse (WOAH). Association for Computational Linguistics.

• Gavin Abercrombie, Dirk Hovy and Vinodkumar Prabhakaran. 2023. Temporal and Second Language Influence on Intra-Annotator Agreement and Stability in Hate Speech Labelling. In Proceedings of LAW-XVII. Association for Computational Linguistics.

• Leonardelli, E., Abercrombie, G., Almanea, D., Basile, V., Fornaciari, T., Plank, B., Poesio, M. and Rieser, V. and Uma, A. 2023. SemEval-2023 Task 11: Learning With Disagreements (LeWiDi). In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023), Association for Computational Linguistics.

• Vitsakis, N., Parekh, A., Dinkar, T., Abercrombie, G., Konstas, I. and Rieser, V. 2023. iLab at SemEval-2023 Task 11 Le-Wi-Di: Modelling Disagreement or Modelling Perspectives? In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023). Association for Computational Linguistics.

• Gavin Abercrombie and Verena Rieser. 2022. Risk-graded Safety for Handling Medical Queries in Conversational AI. In Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing.

• A. Stevie Bergman, Gavin Abercrombie, Shannon Spruit, Dirk Hovy, Emily Dinan, Y-Lan Boureau, and Verena Rieser. 2022. Guiding the Release of Safer E2E Conversational AI through Value Sensitive Design. In Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 39–52, Edinburgh, UK. Association for Computational Linguistics.

• Gavin Abercrombie, Valerio Basile, Sara Tonelli, Verena Rieser and Alexandra Uma. 2022. Proceedings of the First Workshop on Perspectivist Approaches to Natural Language Processing, Marseille, France. European Language Resources Association (ELRA).

• Emily Dinan, Gavin Abercrombie, A. Bergman, Shannon Spruit, Dirk Hovy, Y-Lan Boureau, and Verena Rieser. 2022. SafetyKit: First Aid for Measuring Safety in Open-domain Conversational Systems. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4113–4133, Dublin, Ireland. Association for Computational Linguistics.

• Shiramizu, V. M., Lee, A. J., Altenburg, D., Feinberg, D. R., & Jones, B. C. (2022, March 30). The role of valence, dominance, and pitch in social perceptions of artificial intelligence (AI) conversational agents’ voices. 

• Gavin Abercrombie, Amanda Cercas Curry, Mugdha Pandya, and Verena Rieser. 2021. Alexa, Google, Siri: What are Your Pronouns? Gender and Anthropomorphism in the Design and Perception of Conversational Assistants. In Proceedings of the 3rd Workshop on Gender Bias in Natural Language Processing, pages 24–33, Online. Association for Computational Linguistics.

• Amanda Cercas Curry, Gavin Abercrombie, and Verena Rieser. 2021. ConvAbuse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational AI. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 7388–7403, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.

• Amanda Cercas Curry, Judy Robertson, and Verena Rieser. 2020. Conversational Assistants and Gender Stereotypes: Public Perceptions and Desiderata for Voice Personas. In Proceedings of the Second Workshop on Gender Bias in Natural Language Processing, pages 72–78, Barcelona, Spain (Online). Association for Computational Linguistics.

Science Communication and outreach

• Edinburgh International Science Festival 2022. Drawing activity for children exploring design of conversational assistants.

• West Lothian Libraries. 2021. Gavin Abercrombie: Understanding Online Abuse: an Artificial Intelligence Challenge 

• Edinburgh International Science Festival 2021. Gendering AI: The Case of Conversational Assistants. Selected for Best of the Festival.

Media

• 'Mirages of humanity’: the problem with conversational AI InterMEDIA. 2023 

• Alexa, what are you thinking about? Hello World. 2023.

• Before a Bot Steals Your Job, It Will Steal Your Name The Atlantic. 2023.

• Designing conversational assistants to reduce gender bias  UK Research & Innovation. 2022.

• Why your voice assistant might be sexist BBC Future. 2022.

• Hey Siri, why are all voice assistants female? New research aims to find out The Herald. 2020.