Journal of Advances in Developmental Research

E-ISSN: 0976-4844   •   Impact Factor: 9.71

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Call for Paper Volume 17 Issue 2 July-December 2026 Submit your research before last 3 days of December to publish your research paper in the issue of July-December.
Submit Research Paper Join as a Reviewer

Plagiarism is checked by the leading plagiarism checker TurnItIn

Call for Paper

Volume 17 Issue 2
2026

Submit your research paper

Indexing Partners
Academia Advanced Sciences Index Bielefeld Academic Search Engine CiteSeer DRJI Google Scholar Independent Search Engine & Directory Network ISI (International Scientific Indexing) Issuu Mendeley Research Networks
RefSeek ResearcherId - Thomson Reuters ResearchGate Scirus Scribd Semantic Scholar UTeM - Universiti Teknikal Malaysia Melaka Wiki for Call for Papers WorldCat Zenodo

SENTINEL: Retrieval-Time Governance for Enterprise RAG A Controlled Red-Team Benchmark for Policy-Aware Retrieval, Privacy Enforcement, and Audit Replay

Author(s) Sandeep Nutakki
Country United States
Abstract Retrieval-augmented generation (RAG) systems expose proprietary documents through embedding indexes, rerankers, prompts, and generated answers, yet many enterprise deployments govern only the source repository or redact final text after generation. This paper presents SENTINEL, a governance and privacy architecture for enforcing authorization, purpose limitation, retention, legal hold, and sensitive-data controls inside the retrieval path. SENTINEL combines policy-aware retrieval, embedding provenance, typed sensitive-data classification, prompt-injection filtering, and an append-only audit ledger. We evaluate SENTINEL on a standalone benchmark spanning 1.2 million policy-labeled retrieval requests, 480,000 documents, 12 sensitive-data types, 7 policy classes, and 14 red-team attack families across supply-chain, healthcare, and financial workflows. In this controlled benchmark, SENTINEL reduced unauthorized context exposure from 3.84% under ungoverned dense retrieval to 0.00% observed events (one-sided 95% CI upper bound 0.00025%, McNemar p < 0.001). It achieved 99.2% sensitive-data leakage recall and 97.4% precision, blocked 98.6% of prompt-injection exfiltration attempts, and added 38 ms p95 retrieval latency over a policy-free vector-search baseline. The results support retrieval-time governance as a measurable control for enterprise RAG systems evaluated under known policy labels and red-team attacks, not as evidence of regulatory certification or universal protection.
Keywords retrieval-augmented generation, enterprise RAG, data governance, privacy engineering, access control, prompt injection, auditability, policy-aware retrieval, red-team evaluation
Field Engineering
Published In Volume 17, Issue 1, January-June 2026
Published On 2026-06-25
DOI https://doi.org/10.71097/IJAIDR.v17.i1.2041
View / Download PDF File

Share this



Crossref

Crossref DOI prefix of IJAIDR is 10.71097/IJAIDR

Downloads
Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 17 Isu 1Cover Page Vol 16 Isu 2Cover Page Vol 16 Isu 1

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA Open Access

About IJAIDR
Fees & Payment
Current Issue
Publication Archive 		Submit Research Paper
Track Submission Status
Publication Guidelines
Peer Review & Plagiarism 		Join as a Reviewer
Editors & Reviewers
Get Reviewer Membership Certi. 		Website/Journal Policies
Usage Policy
Content Policies
Privacy Policy
Contact Us Message on WhatsApp +91-9687-182-185 editor@ijaidr.com