Research Outputs #
The following outputs have been produced over the course of the project. The list will continue to be updated as further work is published.
Book chapters / journal articles (peer reviewed and accepted for publication) #
- Berman, Glen, and James Smithies. “Contextualizing AI.” In Operationalizing Responsible AI, edited by Daniele Quercia and Marios Constantinides. Cambridge: Cambridge University Press, forthcoming.
- Smithies, James, Glen Berman, and Karaitiana Taiuru. “Operationalisation of Responsible AI in the Research and Cultural Sectors.” In Building Digital Humanities, edited by Simon Burrows et al. Nijmegen: Radboud University Press, forthcoming.
- Smithies, James, Glen Berman, Karaitiana Taiuru, Martin Spychal, and John Moore. “ATLAS: Harnessing Retrieval Augmented Generation.” In Cambridge Forum on AI: Culture and Society, forthcoming.
Book chapters / journal articles (currently under review) #
- Berman, Glen, and James Smithies. “Epistemic Extractivism in GenAI.” Manuscript submitted for review, n.d.
- McGillivray, Barbara, Sean Lyons, Elizabeth L. Smith, James Smithies, and Ioannis Votsis. “Darwin’s Language Evolution: Decomposing Computational Methods in the Age of AI.” Manuscript submitted for review, n.d.
Conference papers #
- Smithies, James, Glen Berman, Karaitiana Taiuru, Martin Spychal, John Moore, Barbara McGillivray, and Elizabeth Smith. “ATLAS: An Open Test Harness to Enable Transparent Engagement with AI.” Paper presented at DH2026, July 2026 (forthcoming).
- Smithies, James, et al. “ATLAS: Enabling Open Methods and Technical Transparency in AI.” Paper presented at Digital Humanities Australasia, December 4, 2025.
- Berman, Glen, James Smithies, Karaitiana Taiuru, John Moore, Barbara McGillivray, and Martin Spychal. “Evaluation Everywhere, But Not All At Once.” Paper presented at Fantastic Futures 2025, December 5, 2025.
- Smithies, James, Glen Berman, and Karaitiana Taiuru. “Principles for the Operationalisation of Responsible AI in the Research and Public Sector.” Paper presented at eResearch Australasia 2025, October 22, 2025.
- Smithies, James, and Karaitiana Taiuru. “Large Language Models and Transnational Research: Introducing the AI as Infrastructure (AIINFRA) Project.” Paper presented at Fantastic Futures 2024, October 17, 2024.
Invited talks #
- Berman, Glen, and James Smithies. “Reflections on Building ATLAS, a Prototype Open-Source Harness for RAG Evaluation.” The British Library, March 24, 2026.
- Berman, Glen. “Can AI Be Contextualised? Developing a High-Level Framework for Evaluating General Purpose AI for Context Specific Use.” Nokia Bell Labs, November 18, 2025.
- Smithies, James. “Harnessing AI: Evaluating LLM-RAG Systems for Humanities Research.” UC Santa Barbara Transcriptions Centre, May 22, 2025.
Other materials and technical repositories #
- Berman, Glen, and James Smithies. “Responsible AI Principles and the GLAM Sector: An Initial Review.” Zenodo, December 1, 2025. https://doi.org/10.5281/zenodo.17776155.
- Smithies, James, and Glen Berman. ATLAS: Analysis and Testing of Language Models for Archival Systems. Version 0.2.0. GitHub repository, 2025. https://github.com/AI-as-Infrastructure/aiinfra-atlas/releases/tag/v0.2.0.
- Smithies, James, and Glen Berman. “The HASTRICT Principles: A Brief Overview.” Zenodo, October 16, 2025. https://doi.org/10.5281/zenodo.17374271.
- Smithies, James, and Glen Berman. ATLAS Darwin. Version 0.2.0. GitHub repository, 2025. https://github.com/AI-as-Infrastructure/aiinfra-atlas-darwin/releases/tag/v0.2.0.
- Smithies, James, Glen Berman, Barbara McGillivray, and Elizabeth Smith. “ATLAS Darwin Telemetry 06-10-2025.” Zenodo dataset, October 31, 2025. https://doi.org/10.5281/zenodo.17491578.
- Smithies, James. New Zealand Hansard Data Preparation. GitHub repository, 2024. https://github.com/AI-as-Infrastructure/aiinfra-nzhansard-preparation.