Metraipolis › archive › dc_fa152716a294
Specification-Driven Governance for Fairness-by-Design in Agentic Software Engineering: How Far Can We Go? — arXiv:2611.03231
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.03231
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:41:44.363Z
- SHA-256 of the archived text
b59b8d9d303f430536d90b4f3670a15865c80b2752a8fe4c6ce483000141870f- Stable URL of this page
https://metraipolis.com/documento/dc_fa152716a294
How to cite
Specification-Driven Governance for Fairness-by-Design in Agentic Software Engineering: How Far Can We Go? — arXiv:2611.03231 — arXiv (en), https://arxiv.org/abs/2611.03231 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fa152716a294 (Metraipolis archive dc_fa152716a294, sha256 b59b8d9d303f430536d90b4f3670a15865c80b2752a8fe4c6ce483000141870f)
BibTeX
@misc{dcfa152716a294,
title = {Specification-Driven Governance for Fairness-by-Design in Agentic Software Engineering: How Far Can We Go? — arXiv:2611.03231},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.03231},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fa152716a294; sha256 b59b8d9d303f430536d90b4f3670a15865c80b2752a8fe4c6ce483000141870f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03231. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Agentic AI is shifting developers' responsibilities from software implementation to software governance. While this paradigm improves productivity, it also raises challenges in ensuring that autonomous coding agents preserve developers' intentions. Among these, operationalizing fairness-by-design is largely unexplored, as existing approaches provide limited support for representing and propagating fairness requirements throughout agentic development workflows. In this paper, we investigate whether specification-driven development can operationalize fairness-by-design in the agentic development of machine learning systems. We study three coding agents on four machine learning applications, using four prompts that progressively operationalize fairness through nine Fairness Specification Elements, and analyze the traceability of fairness requirements across the generated artifacts, the fairness-accuracy trade-off of the resulting models, and the agents' design decisions and rationales. The results show that when fairness is operationalized, agents capture almost all fairness requirements in the specification (97.22%) and preserve most of them in the code (82.41%). However, the resulting models are not significantly fairer, as agents implement fairness as a post-hoc audit that neither shapes training nor prevents model delivery. Specifications thus make fairness traceable, but not enforceable, calling for fairness specifications that state what agents should do when requirements are not met, and for workflows that make the fairness decisions agents delegate to developers explicit. Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.SE, nuovi lavori del giorno). Categorie arXiv: Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2611.03231 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T18:13:07.763Z dalla pagina https://arxiv.org/list/cs.SE/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).