EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Opportunities for Ethical AI Integration in Core Infrastructure Administration: Safety Boundaries and Control Frameworks for Autonomous DBAs

Evans Ighodalo, Ayinoluwa Kolawale, Chukwudera Obumneke Anunagba, Oluranti Ogundapo, Harouna Wendpanga Yann Christian Sankara, Emmanuel Chukwuma Okeke, Toussida Fatah Tanguy Minoungou, Aisha Ife Abdulwahab

Journal of Emerging Intelligence and Engineering Technologies · 10.11.2026 · 3 (02) · 113-151

DOI: 10.66592/mduc9e ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Emerging Intelligence and Engineering Technologies, online erschienen am 10.11.2026. Verfasst von Evans Ighodalo und 7 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Database administration is one of the oldest and most consequential disciplines in enterprise computing. Databases hold the records on which banks, hospitals, utilities, governments and commercial platforms depend, and the people who administer them have traditionally held some of the most powerful privileges in any organisation. Between 2017 and 2026 the discipline moved from rule-based self-tuning features toward agentic systems built on large language models (LLMs) that can diagnose incidents, rewrite queries, apply patches and change schemas with little or no human involvement. Major vendors now market platforms that observe, reason and act on production databases, and practitioners increasingly describe the role of the human DBA as one of orchestrating AI agents rather than executing scripts. This paper examines the opportunities that autonomous database administration offers for core infrastructure and the ethical, operational and regulatory risks it introduces. Methodologically, it combines an integrative literature synthesis of 259 sources, including 238 seed studies screened from a consolidated multi-portfolio corpus of 1,056 records, with a design science approach to building and analytically evaluating a conceptual model. Drawing on the documented July 2025 incident in which an AI coding agent deleted a live production database during a declared code freeze, the OWASP Top 10 for Agentic Applications, the European Union Artificial Intelligence Act and the United States NIST AI Agent Standards Initiative, the paper argues that the central design problem is not whether AI should administer databases but how authority should be bounded, delegated, observed and revoked. It proposes three contributions: (1) a taxonomy of database actions classified by reversibility and blast radius; (2) a six-level Autonomy Ladder for Database Operations (ALDO) that ties levels of independence to evidence of reliability; and (3) a Layered Control Framework comprising policy, identity, execution, verification and accountability layers. These elements are integrated into a conceptual model with five propositions. The paper discusses the model against the literature and the regulatory environment, sets out theoretical, practical and policy implications, and closes with limitations and an agenda for future research

Zitieren

APA

Ighodalo, E., Kolawale, A., Anunagba, C. O., Ogundapo, O., Sankara, H. W. Y. C., Okeke, E. C., Minoungou, T. F. T., & Abdulwahab, A. I. (2026). Opportunities for Ethical AI Integration in Core Infrastructure Administration: Safety Boundaries and Control Frameworks for Autonomous DBAs. Journal of Emerging Intelligence and Engineering Technologies, 3(02), 113-151. https://doi.org/10.66592/mduc9e

Vancouver

Ighodalo E, Kolawale A, Anunagba CO, Ogundapo O, Sankara HWYC, Okeke EC, et al. Opportunities for Ethical AI Integration in Core Infrastructure Administration: Safety Boundaries and Control Frameworks for Autonomous DBAs. Journal of Emerging Intelligence and Engineering Technologies. 2026;3(02):113-151. doi:10.66592/mduc9e

BibTeX
@article{Ighodalo2026,
  author = {Ighodalo, Evans and Kolawale, Ayinoluwa and Anunagba, Chukwudera Obumneke and Ogundapo, Oluranti and Sankara, Harouna Wendpanga Yann Christian and Okeke, Emmanuel Chukwuma and Minoungou, Toussida Fatah Tanguy and Abdulwahab, Aisha Ife},
  title = {Opportunities for Ethical AI Integration in Core Infrastructure Administration: Safety Boundaries and Control Frameworks for Autonomous DBAs},
  journal = {Journal of Emerging Intelligence and Engineering Technologies},
  year = {2026},
  volume = {3},
  number = {02},
  pages = {113--151},
  doi = {10.66592/mduc9e}
}
Kostenlos starten →