DPhil Social Data Science Candidate working on the fairness of AI Agents.

DPhil Poster Presentation (Michaelmas 2024)

Date:

Below are the references and resources used for my DPhil poster presentation at the Oxford Internet Institute.

References

Why care about AI agents?

Chan, A., Salganik, R., Markelius, A., Pang, C., Rajkumar, N., Krasheninnikov, D., Langosco, L., He, Z., Duan, Y., Carroll, M., Lin, M., Mayhew, A., Collins, K., Molamohammadi, M., Burden, J., Zhao, W., Rismani, S., Voudouris, K., Bhatt, U., … Maharaj, T. (2023). Harms from increasingly agentic algorithmic systems. Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency, 651–666. https://doi.org/10.1145/3593013.3594033

Eloundou, T., Manning, S., Mishkin, P., & Rock, D. (2024). GPTs are GPTs: Labor market impact potential of LLMs. Science, 384(6702), 1306–1308. https://doi.org/10.1126/science.adj0998

Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., Zhao, W. X., Wei, Z., & Wen, J. (2024). A survey on large language model based autonomous agents. Frontiers of Computer Science, 18(6), 186345. https://doi.org/10.1007/s11704-024-40231-1 f

Why the Danish public sector?

Ilsøe, A., Larsen, T. P., Mathieu, C., & Rolandsson, B. (2024). Negotiating about Algorithms: Social Partner Responses to AI in Denmark and Sweden. ILR Review, 77(5), 856–868. https://doi.org/10.1177/00197939241278956f

Jørgensen, R. F. (2023). Data and rights in the digital welfare state: The case of Denmark. Information, Communication & Society, 26(1), 123–138. https://doi.org/10.1080/1369118X.2021.1934069

What are current limitations

Kapoor, S., Stroebl, B., Siegel, Z. S., Nadgir, N., & Narayanan, A. (2024). AI agents that matter (No. arXiv:2407.01502). arXiv. https://doi.org/10.48550/arXiv.2407.01502

Selbst, A. D., boyd, danah, Friedler, S. A., Venkatasubramanian, S., & Vertesi, J. (2019). Fairness and Abstraction in Sociotechnical Systems. Proceedings of the Conference on Fairness, Accountability, and Transparency, 59–68. https://doi.org/10.1145/3287560.3287598

Weidinger, L., Rauh, M., Marchal, N., Manzini, A., Hendricks, L. A., Mateos-Garcia, J., Bergman, S., Kay, J., Griffin, C., Bariach, B., Gabriel, I., Rieser, V., & Isaac, W. (2023). Sociotechnical Safety Evaluation of Generative AI Systems (No. arXiv:2310.11986). arXiv. http://arxiv.org/abs/2310.11986

Paper 1: Political bias

Bolukbasi, T., Chang, K.-W., Zou, J. Y., Saligrama, V., & Kalai, A. T. (2016). Man is to computer programmer as woman is to homemaker? Debiasing word embeddings. Advances in Neural Information Processing Systems, 29, 4349–4357.

Gordon, J., Babaeianjelodar, M., & Matthews, J. (2020). Studying Political Bias via Word Embeddings. Companion Proceedings of the Web Conference 2020, 760–764. https://doi.org/10.1145/3366424.3383560

Rystrøm, J. (2024). Jhrystrom/wicked-fair [Computer software]. https://github.com/jhrystrom/wicked-fair (Original work published 2024)

Tessler, M. H., Bakker, M. A., Jarrett, D., Sheahan, H., Chadwick, M. J., Koster, R., Evans, G., Campbell-Gillingham, L., Collins, T., Parkes, D. C., Botvinick, M., & Summerfield, C. (2024). AI can help humans find common ground in democratic deliberation. Science, 386(6719), eadq2852. https://doi.org/10.1126/science.adq2852

Paper 2: Fairness in tool use

Ye, J., Li, S., Li, G., Huang, C., Gao, S., Wu, Y., Zhang, Q., Gui, T., & Huang, X. (2024). ToolSword: Unveiling Safety Issues of Large Language Models in Tool Learning Across Three Stages. In L.-W. Ku, A. Martins, & V. Srikumar (Eds.), Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 2181–2211). Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.acl-long.119

Paper 3: Fair agent framework

Mökander, J., Schuett, J., Kirk, H. R., & Floridi, L. (2024). Auditing large language models: A three-layered approach. AI and Ethics, 4(4), 1085–1115. https://doi.org/10.1007/s43681-023-00289-2

Resources

The poster used beautiful icons from the Noun Project. Credits are below 👇

  • Public Service by putri amaliya from Noun Project (CC BY 3.0)
  • robot software by Dava Arya Ditya from Noun Project (CC BY 3.0)
  • Tool by Suminah Wulandari from Noun Project (CC BY 3.0)
  • bias by Nithinan Tatah from Noun Project (CC BY 3.0)
  • Miscommunication by Prashanth Rapolu from Noun Project (CC BY 3.0)
  • government digital service by Eucalyp from Noun Project (CC BY 3.0)
  • Culture by Chaiwat Kinkaew from Noun Project (CC BY 3.0)