Home / Current Issue / Paper 1710555
Designing Adaptive AI Models for Proactive Privacy Risk Anticipation in Cloud-Based Infrastructures
Subject area: Science,Engineering and Technology · Area of research: Artificial Intelligence
Abstract
Cloud computing has made itself an integral aspect of contemporary digital ecosystems. At the same time, the use of multi-tenancy, dynamic workloads, and cross-border data transfers introduce risk to personal privacy. Compliance frameworks are primarily reactive, leaving organizations unguarded to evolving security threats that violate privacy. This thesis looks into the preliminary design of AI models that may self-learn and anticipate privacy invasion risks within an infrastructure that is hosted on the cloud. Altogether, 250 events that simulate the invasion of privacy in cloud computing systems were used to test three adaptive components: predictive analytics, reinforcement learning, and anomaly detection. The results show adaptive AI?s ability to recognize and track privacy breaches in addition to lowering false positive alerts and bolstering defenses against emerging threats. The knowledge of privacy protection that is AI-driven has certainly grown, and the concrete steps explored in this thesis are emerging to streamline real-world adaptive compliance systems in cloud settings.
Keywords
AI, AI model, Cloud Based, Infrastructure, Risk
How to cite this paper
@article{1710555,
author = {Jennifer Olomina},
title = {Designing Adaptive AI Models for Proactive Privacy Risk Anticipation in Cloud-Based Infrastructures},
journal = {Iconic Research And Engineering Journals},
year = {2024},
volume = {8},
number = {1},
pages = {753-758},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1710555.pdf},
abstract = {Cloud computing has made itself an integral aspect of contemporary digital ecosystems. At the same time, the use of multi-tenancy, dynamic workloads, and cross-border data transfers introduce risk to personal privacy. Compliance frameworks are primarily reactive, leaving organizations unguarded to evolving security threats that violate privacy. This thesis looks into the preliminary design of AI models that may self-learn and anticipate privacy invasion risks within an infrastructure that is hosted on the cloud. Altogether, 250 events that simulate the invasion of privacy in cloud computing systems were used to test three adaptive components: predictive analytics, reinforcement learning, and anomaly detection. The results show adaptive AI?s ability to recognize and track privacy breaches in addition to lowering false positive alerts and bolstering defenses against emerging threats. The knowledge of privacy protection that is AI-driven has certainly grown, and the concrete steps explored in this thesis are emerging to streamline real-world adaptive compliance systems in cloud settings.},
keywords = {AI, AI model, Cloud Based, Infrastructure, Risk},
month = {July},
}