The Paradox of Personalization: A Zero-Trust Architecture for AI-Driven CX Platforms in Regulated Multi-Cloud Environments

Approved

Relevance and Originality

The manuscript addresses an important challenge in financial technology by combining personalized services with privacy and regulatory requirements. The integration of homomorphic encryption and zero trust principles is relevant, but the authors should clarify the specific research gap and distinguish the proposed approach from existing privacy preserving AI frameworks.

Methodology

The architecture is explained with useful technical details, including encryption, validation, and key management mechanisms. However, more information about the implementation environment, datasets, evaluation methods, and system configuration is needed to improve reproducibility.

Validity and Reliability

The validation results indicate improvements in performance and compliance. The manuscript would benefit from clearer details on baseline comparisons, testing conditions, workload characteristics, and limitations related to scalability and practical deployment.

Clarity and Structure

The manuscript follows a logical structure and presents technical concepts effectively. Some sections require clearer explanations and smoother connections between architectural design and evaluation results to improve accessibility.

Results and Analysis

The findings show potential benefits for privacy preserving customer experience systems. A stronger comparison with alternative approaches and a deeper discussion of security, performance, and operational tradeoffs would improve the analysis.