Computational hermeneutics: evaluating generative AI as a cultural technology
Generative AI (GenAI) systems are increasingly recognized as cultural technologies, yet current evaluation frameworks often treat culture as a variable to be measured rather than fundamental to the system's operation. Drawing on hermeneutic theory from the humanities, we argue that GenAI systems function as "context machines" that must inherently address three interpretive challenges: situatedness (meaning only emerges in context), plurality (multiple valid interpretations coexist), and ambiguity (interpretations naturally conflict). We present computational hermeneutics as an emerging framework offering an interpretive account of what GenAI systems do, and how they might do it better. We offer three principles for hermeneutic evaluation-that benchmarks should be iterative, not one-off; include people, not just machines; and measure cultural context, not just model output. This perspective offers a nascent paradigm for designing and evaluating contemporary AI systems: shifting from standardized questions about accuracy to contextual ones about meaning.
Authors
- Maria Antoniak (ORCID: https://orcid.org/0000-0002-4807-2850)
- Hoyt Long (ORCID: https://orcid.org/0000-0002-8562-5426)
- Shauna Concannon (ORCID: https://orcid.org/0000-0002-5904-0045)
- Drew Hemment (ORCID: https://orcid.org/0000-0002-0068-5500)
- Martin Disley (ORCID: https://orcid.org/0000-0001-7768-7371)
- Matthew Wilkens (ORCID: https://orcid.org/0000-0001-6749-9318)
- Kerry Francksen (ORCID: https://orcid.org/0009-0006-3454-5537)
- A.J. Sobey (ORCID: https://orcid.org/0000-0001-6880-8338)
- E. Powell Robinson
- Richard Jean So (ORCID: https://orcid.org/0009-0002-6442-055X)
- Deven M. Parker
- Kirsten Ostherr (ORCID: https://orcid.org/0000-0002-6425-9527)
- T. Underwood
- James E. Dobson (ORCID: https://orcid.org/0000-0001-8357-7240)
- Jessica Ratcliff
- Cody Kommers
- Ryan Heuser (ORCID: https://orcid.org/0000-0002-4155-4961)
- Emmanouil Benetos
- Ashley Noel-Hirst (ORCID: https://orcid.org/0009-0001-7987-9905)
- Yali Du
- Yipeng Qin
- J.A. Gray (ORCID: https://orcid.org/0000-0002-7287-0748)
- S. Rebecca Leigh
- Steve A. Benford
- Georgia A. Meyer
- Mercedes Bunz
- Sarah Immel
- Ruth Ahnert
- Dalaki Livingston
- Aditya Vashistha
- D. Mihai
- Evelyn Gius
- Yuan Zheng
- Karina Rodriguez
- Meredith Martin
- Edgar Duéñez-Guzmán
- Youyou Wu
Institutions
- Dartmouth College (US)
- American Institute of Biological Sciences (US)
- University of Nottingham (GB)
- University of Illinois Urbana-Champaign (US)
- Queen Mary University of London (GB)
- King's College London (GB)
- Durham University (GB)
- Princeton University (US)
- Cornell University (US)
- University of Utah (US)
- University of Exeter (GB)
- University of Cambridge (GB)
- University of Colorado System (US)
- Technische Universität Darmstadt (DE)
- Sorbonne Université (FR)
- University of Chicago (US)
- The Alan Turing Institute (GB)
- Dartmouth Hospital (GB)
- University of Southampton (GB)
- Laboratoire Médiations (FR)
- University College London (GB)
- McGill University (CA)
- University of Brighton (GB)
- Coventry University (GB)
- Rice University (US)
- University of Glasgow (GB)
- Cardiff University (GB)
- London School of Economics and Political Science (GB)
- University of Sheffield (GB)
- University of Edinburgh (GB)
Publication Details
- Journal
- Frontiers in Artificial Intelligence
- Published
- 2026-02-26
- DOI
- https://doi.org/10.3389/frai.2026.1753041
- Primary Topic
- Computational and Text Analysis Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Lloyd's Register
- Arts and Humanities Research Council