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 1 January-June 2026 Submit your research before last 3 days of June to publish your research paper in the issue of January-June.
Submit Research Paper Join as a Reviewer

Plagiarism is checked by the leading plagiarism checker TurnItIn

Call for Paper

Volume 17 Issue 1
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

HADES: Hallucination-Aware Detection and Evaluation System for Large Language Model Outputs in Enterprise CRM Workflows

Author(s) Lalith Chandra Bandaru
Country United States
Abstract Large language models integrated into enterprise Salesforce CRM workflows for automated email drafting, lead qualification, knowledge article generation, and case resolution introduce a distinctive business risk: LLM hallucination, in which fluent, professionally formatted outputs contain factually incorrect or fabricated claims that business users accept without the source verification expertise needed to identify the inaccuracy. In CRM contexts, hallucinations can produce customer-facing communications with incorrect pricing, invented product specifications, misattributed customer preferences, or unsourced regulatory claims — each with potential for contractual, reputational, or compliance consequences. HADES (Hallucination-Aware Detection and Evaluation System) addresses this risk through a three-layer evidence architecture that evaluates each factual claim in an LLM output against the CRM record context, an organisational knowledge base, and a factual verification index, assigning a confidence score to each claim and routing low-confidence outputs to human review while certifying high-confidence outputs for direct use. Evaluated across six enterprise Salesforce deployments over twelve months covering 1.2 million LLM outputs, HADES achieved 94.2% precision and 89.7% recall for hallucination detection through a gradient-boosted ensemble of three verification components, reduced human review volume by 73% compared to universal manual oversight, and contributed to an 81% reduction in hallucination-attributable CRM incidents.
Keywords LLM hallucination, enterprise AI, Salesforce, CRM, factual verification, retrieval-augmented generation, AI safety, claim extraction, confidence scoring, human-in-the-loop.
Field Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In Volume 14, Issue 1, January-June 2023
Published On 2023-05-20
DOI https://doi.org/10.71097/IJAIDR.v14.i1.1964
Short DOI https://doi.org/hb5p9t
View / Download PDF File

Share this



CrossRef

CrossRef DOI is assigned to each research paper published in our journal.

IJAIDR DOI prefix 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