EUVIMEDEUVIMED entdecken
Crossref

AI-Driven Risk-Adaptive Authorization for Multi-Tenant Cloud APIs - Microsoft Azure

Balaji Chode

International Journal of Computer Science and Data Engineering · 2025

DOI: 10.55124/csdb.v2i3.254 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Science and Data Engineering, 2025. Verfasst von Balaji Chode.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Application Programming Interfaces (APIs) have become the nervous system of modern financial-services platforms; yet fractured authorization logic remains a dominant breach vector, exposing organizations to data exfiltration, fraudulent transactions and regulatory fines. We introduce a cloud-native Contextual Authorization Framework (CAF) embedded in an enterprise Shared Services Platform (SSP) that supports more than 200 customers facing and internal micro-services worldwide. CAF sits behind Azure API Management, authenticates callers via OAuth 2.0 / OpenID Connect, and merges an ensemble machine learning risk score—Isolation Forest, GRU auto-encoder and XGBoost—with attribute-based rules expressed as policy-as-code in Open Policy Agent. A twelve-month evaluation covering 2.1 billion production requests demonstrates that CAF increases attack-detection recall by 42 % and precision by 18 % compared with a signature Web Application Firewall and static RBAC baseline, while adding only 8 ms to the p95 gateway latency—well inside the 15 ms service-level objective required for real-time quote, billing and claim APIs. Operational metrics show a 60 % reduction in security-integration effort and a net annual benefit of $7.8 million due to prevented fraud and lower SOC triage workload. We have proven datasets, Azure ML notebooks and policy templates, demonstrating that latency-bounded, AI-augmented authorization is both technically feasible and economically compelling for enterprises pursuing Zero-Trust maturity. Keywords: API security, Zero Trust Architecture, Contextual Authorization, Cloud-native access control, Policy-as-Code, Open Policy Agent (OPA), OAuth 2.0, OpenID Connect, Machine Learning for security, Risk-adaptive authorization, XGBoost, GRU auto encoder, Isolation Forest, Azure API Management, Real-time access control, Shared Services Platform, Financial services cyber security, AI-driven fraud prevention, Authorization latency optimization, Security integration automation

Zitieren

APA

Chode, B. (2025). AI-Driven Risk-Adaptive Authorization for Multi-Tenant Cloud APIs - Microsoft Azure. International Journal of Computer Science and Data Engineering. https://doi.org/10.55124/csdb.v2i3.254

Vancouver

Chode B. AI-Driven Risk-Adaptive Authorization for Multi-Tenant Cloud APIs - Microsoft Azure. International Journal of Computer Science and Data Engineering. 2025. doi:10.55124/csdb.v2i3.254

BibTeX
@article{Chode2025,
  author = {Chode, Balaji},
  title = {AI-Driven Risk-Adaptive Authorization for Multi-Tenant Cloud APIs - Microsoft Azure},
  journal = {International Journal of Computer Science and Data Engineering},
  year = {2025},
  doi = {10.55124/csdb.v2i3.254}
}
Kostenlos starten →