Page 381
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
AI and Zero-Trust Architecture for Securing Data in Remote Work
Settings: A Comparative Study
Marianne Ghilyn V. Golo, Eduardo R. Yu II, Reagan B. Ricafort
AMA University
DOI:
https://doi.org/10.51583/IJLTEMAS.2026.150400034
Received: 10 April 2026; Accepted: 15 April 2026; Published: 05 May 2026
ABSTRACT
The COVID19 pandemic accelerated remote and hybrid work adoption, exposing organizations to insider
threats, data breaches, and advanced cyberattacks, which traditional perimeter-based models failed to address;
in response, Zero Trust Architecture (ZTA) emerged, and its integration with Artificial Intelligence (AI) has
become a cornerstone of cybersecurity strategies by enabling anomaly detection, automated policy enforcement,
and rapid incident response. Guided by PRISMA methodology and Rapid review principles, this study
systematically examined 25 publications from 2020–2030 across IEEE Xplore, ACM Digital Library, MDPI,
SpringerLink, Elsevier, government repositories, and open access archives, applying strict eligibility criteria to
ensure methodological transparency and relevance. Findings consistently show that AI-ZTA integration
mitigates insider threats, prevents data breaches, and strengthens resilience against advanced cyberattacks, with
chronological analysis revealing a progression from foundational frameworks (2020–2023), to risk-oriented
literature (2024), applied deployments (2024–2025), and predictive analyses (2025–2026). The review
concludes that AI-ZTA is positioned as a critical paradigm for securing decentralized environments, though its
long-term success depends on safeguards, workforce training, regulatory compliance, and continuous evaluation
mechanisms. This scope and format are consistent with established practices in cybersecurity research, where
recent studies have also synthesized fewer than 25 papers through rapid review methods to deliver timely,
rigorous, and actionable insights in emerging fields.
INTRODUCTION
The COVID-19 pandemic accelerated the adoption of remote and hybrid work models, reshaping organizational
operations and exposing companies to heightened cybersecurity risks. Peer-reviewed studies indicate that more
than 60% of organizations reported remote work-related security incidents, with remote employees being about
twice as likely to fall victim to phishing attacks compared to on-site staff (Sabin, 2021; Nizamuddin, 2025). The
average cost of a data breach in remote work contexts has exceeded $4 million globally per incident,
underscoring the financial and operational impact of inadequate security frameworks (Nizamuddin, 2025). At
the same time, projections suggest that cybercrime costs will continue to escalate sharply through 2030, driven
by ransomware, phishing, and AI-enabled attacks (Sabin, 2021).
Traditional perimeter-based security models, designed for centralized office environments, proved inadequate
in addressing insider threats, data breaches, and advanced cyberattacks. In response, Zero-Trust Architecture
(ZTA), based on the principle of “never trust, always verify”—emerged as a transformative cybersecurity
framework (Rose et al., 2020; Gambo & Almulhem, 2025). More recently, the integration of AI into ZTA has
become a defining trend, enabling anomaly detection, automated policy enforcement, and rapid incident
response. This convergence positions AI-ZTA as a cornerstone of cybersecurity strategies for decentralized
workforces.
Despite its promise, AI-ZTA integration faces several challenges. Studies highlight risks such as the erosion of
Zero-Trust principles through generative AI (Xu et al., 2025), high implementation costs and workforce training
gaps (Rodrigues, 2026), and ethical concerns regarding surveillance and algorithmic decision-making (Gartner,
2026). Moreover, while AI and ZTA have individually demonstrated effectiveness, limited comparative research
Page 382
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
has examined their combined impact in organizational and remote work contexts. This gap prevents
organizations from fully understanding the benefits, drawbacks, and long-term implications of AI-ZTA adoption.
This study aims to conduct a comparative analysis of AI-ZTA integration in securing data within remote work
environments between 2020 and 2030. It seeks to examine how organizations across industries have
implemented AI enhanced Zero-Trust frameworks, to evaluate their effectiveness relative to traditional security
models, and to analyze their role in mitigating insider threats, data breaches, and advanced cyberattacks.
Furthermore, the study traces the chronological progression of AI-ZTA adoption, from foundational frameworks
to applied case studies and predictive analyses, in order to clarify both its technical effectiveness and practical
adaptability. By systematically reviewing studies published between 2020 and 2030, this research contributes
to the literature by bridging the gap between conceptual frameworks (Rose et al., 2020; Ajish, 2024b), applied
case studies (Nzeako & Shittu, 2024; Ajimatanrareje & Agbesi, 2025), and predictive analyses (Xu et al., 2025;
Ucheji, 2026). The findings highlight both the technical effectiveness and practical relevance of AI-ZTA, while
identifying unresolved challenges related to cost, compliance, and governance. Ultimately, this study positions
AI-ZTA as a critical paradigm for future cybersecurity strategies, contingent upon robust safeguards and
continuous evaluation mechanisms.
Research Questions and Objectives
Developing research questions and objectives is essential to carrying out a systematic review because it provides
a clear and deliberate focus that directs processes such as study selection, data extraction, and synthesis. The
following research questions and objectives guide the process of conducting this review on AI-ZTA integration:
Research Questions (RQs):
• RQ1: What are the applications of AI-ZTA across organizational, cloud, and remote workforce contexts?
• RQ2: What challenges and limitations are encountered in the adoption and implementation of AI-ZTA
frameworks?
• RQ3: How effective is AI-ZTA integration in mitigating insider threats, preventing data breaches, and
enhancing resilience against advanced cyberattacks?
• RQ4: What future opportunities and risks are forecasted for AI-ZTA adoption between 2020 and 2030?
Research Objectives (ROs):
• RO1: To identify and analyze applications of AI-ZTA across industries and technological environments.
• RO2: To investigate the key issues and challenges in adopting and implementing AI-ZTA frameworks.
• RO3: To evaluate the effectiveness of AI-ZTA integration in addressing insider threats, data breaches,
and advanced cyberattacks.
• RO4: To trace the chronological progression of AI-ZTA adoption from foundational frameworks to
predictive analyses.
• RO5: To explore future opportunities and risks, including the impact of emerging technologies such as
generative AI on Zero-Trust principles.
AI-ZTA Integration
AI-ZTA integration refers to the convergence of Artificial Intelligence (AI) with Zero-Trust Architecture (ZTA)
to enhance cybersecurity resilience in decentralized environments such as remote work, cloud systems, and
critical infrastructure. ZTA, founded on the principle of “never trust, always verify,” emphasizes continuous
authentication, authorization, and monitoring of users, devices, and applications (Rose et al., 2020; Gambo &
Page 383
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Almulhem, 2025; CISA, 2023). AI strengthens these principles by automating anomaly detection, enforcing
adaptive policies, and enabling rapid incident response, thereby transforming ZTA from a static framework into
a dynamic, proactive defense model (Ajish, 2024b; Paul et al., 2024).
Principles and core elements
The integration of AI-ZTA is composed of several core components. First, continuous authentication and identity
management ensures that access is verified at every stage, with AI detecting anomalies in login behavior and
device usage (Ajish, 2024b). Second, anomaly detection and threat intelligence leverage machine learning to
identify unusual patterns in user activity or network traffic, enabling early detection of insider threats and
advanced persistent attacks (Karamchand, 2024; Paul et al., 2024). Third, automated policy enforcement allows
AI to dynamically adjust access privileges in real time, reducing reliance on manual intervention and
strengthening compliance (Ajish, 2024b; Gadkari, 2025). Fourth, incident response and resilience are improved
as AI accelerates detection and containment of breaches, reducing mean time to respond (Xu et al., 2025; Ucheji,
2026). Fifth, data protection and encryption are reinforced by AI-driven monitoring that prevents unauthorized
data exfiltration in cloud and remote environments (Nzeako & Shittu, 2024; Kodi, 2025). Finally, governance,
risk, and compliance (GRC) are supported through AI-enabled monitoring of regulatory standards, audit trails,
and continuous evaluation mechanisms (Rodrigues, 2026; Gartner, 2026).
These components illustrate how AI-ZTA integration shifts cybersecurity from reactive defense to proactive
resilience, positioning it as a cornerstone of future organizational security strategies. Applied case studies in
cloud and infrastructure contexts (Nzeako & Shittu, 2024; Ajimatanrareje & Agbesi, 2025) further demonstrate
its scalability and adaptability, while predictive analyses highlight both opportunities and risks for future
adoption (Xu et al., 2025; Ucheji, 2026). Table 1 outlines the core components of AI-ZTA integration, showing
how each principle of Zero Trust is strengthened by AI functions.
ZTA Principle
AI Function in Integration
Key Contribution
Continuous
Authentication &
Identity Management
AI detects anomalies in login
behavior, device usage, and access
requests (Ajish, 2024b)
Strengthens adaptive multi-factor
authentication and identity verification (Rose
et al., 2020;
CISA, 2023)
Anomaly Detection &
Threat Intelligence
Machine learning models analyze
user activity and network traffic
(Karamchand, 2024; Paul et al.,
2024)
Enables early detection of insider threats and
advanced persistent attacks
Automated Policy
Enforcement
AI dynamically adjusts access
privileges in real time (Ajish,
2024b; Gadkari, 2025)
Reduces manual intervention and ensures
compliance
Incident Response &
Resilience
AI accelerates breach detection and
containment (Xu et al., 2025;
Ucheji, 2026)
Minimizes mean time to respond
(MTTR) and improves resilience
Data Protection &
Encryption
AI-driven monitoring prevents
unauthorized data exfiltration
(Nzeako & Shittu, 2024; Kodi,
2025)
Secures sensitive data across cloud and
remote environments
Governance, Risk, and
Compliance (GRC)
AI monitors regulatory standards
and generates audit trails
(Rodrigues, 2026; Gartner, 2026)
Supports continuous evaluation and
accountability
Table 1. AI-ZTA Integration: Core Components and Functions.
Page 384
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Comparative analysis of AI–ZTA vs. Traditional ZTA
Traditional ZTA has worked well in centralized office environments, but its reliance on static multi-factor
authentication and manual enforcement limits its effectiveness in today’s decentralized settings (Rose et al.,
2020; CISA, 2023). AI-ZTA builds on this foundation by adding adaptive, automated, and predictive
mechanisms. These enhancements improve anomaly detection, accelerate incident response, and make the
framework more scalable for remote and cloud contexts (Ajish, 2024b; Paul et al., 2024; Karamchand, 2024).
Table 2 highlights these differences across six security criteria.
Criteria
Traditional ZTA
AI–ZTA Integration
Authentication
Static MFA, periodic checks
Continuous, adaptive, anomaly-based
Threat Detection
Rule-based, reactive
Machine learning, proactive
anomaly detection
Policy Enforcement
Manual, predefined
Automated, dynamic, real-time
Incident Response
Human-led, slower MTTR
AI-accelerated, reduced MTTR
Scalability
Limited in complex remote setups
Highly scalable across cloud & remote work
Compliance
Manual audits
AI-driven monitoring, automated audit trails
Table 2. Comparative Analysis of Traditional ZTA and AI-ZTA Integration.
Figure 1 demonstrates that AI-ZTA consistently outperforms traditional ZTA across critical security areas. The
blue bars, representing traditional ZTA, are noticeably shorter, reflecting its limited effectiveness in
authentication, threat detection, policy enforcement, incident response, scalability, and compliance (Rose et al.,
2020; CISA, 2023). In contrast, the orange bars for AIZTA extend further, showcasing stronger performance
enabled by automation, anomaly detection, and adaptive policy enforcement (Ajish, 2024b; Paul et al., 2024;
Karamchand, 2024). Despite these clear advantages, challenges such as higher implementation costs, workforce
training requirements, and ethical concerns about surveillance remain significant considerations (Rodrigues,
2026; Gartner, 2026).
Figure 1. Visual Comparison of Traditional vs. AI-ZTA Across Security Criteria.
Page 385
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Together, Table 2 and Figure 1 show that while traditional ZTA provides the baseline, AI-ZTA elevates security
to a higher level of adaptability and resilience. These improvements reduce human dependency, accelerate
response times, and enhance resilience in remote work environments (Xu et al., 2025; Ucheji, 2026).
Yet, organizations must balance these advantages against trade-offs such as higher implementation costs,
workforce readiness challenges, and ethical concerns about surveillance (Rodrigues, 2026; Gartner, 2026).
METHODOLOGIES
This study used a rapid systematic review to keep the process both transparent and rigorous. The approach
followed the PRISMA framework, which provided a clear structure for identifying, screening, and synthesizing
relevant literature on AI-ZTA integration.
To keep the review focused, the scope was narrowed to 25 studies. This number is consistent with other
cybersecurity reviews and ensures depth without overwhelming breadth. PRISMA guidelines support
transparency by requiring eligibility criteria, flow diagrams, and structured evidence tables.
By combining rapid review techniques with PRISMA standards, the study achieved a balance between speed
and rigor. The result is a methodology that is easy to replicate, reviewer friendly, and well suited to fast-moving
fields like AI-ZTA integration.
Eligibility Criteria
To ensure rigor and relevance, the review applied inclusion and exclusion criteria consistent with the PRISMA
framework. Studies were selected based on parameters informed by established Zero Trust frameworks (Rose
et al., 2020; CISA, 2023), systematic reviews (Ahmad, 2025; Liman Gambo & Almulhem, 2025; Zakhmi et al.,
2025), and recent AI-ZTA integration studies (Ajish, 2024b; Karamchand, 2024; Ucheji, 2026). These references
provided the methodological foundation for defining inclusion and exclusion parameters.
Inclusion criteria
Exclusion criteria
Published between 2020 and 2030 (Rose et al., 2020;
Ajish, 2024b; Ucheji, 2026)
Studies outside the 2020–2030
timeframe
Addressed AI-ZTA integration directly, or provided
foundational ZTA frameworks, AI trust/security
studies, systematic reviews, or government/industry
guidelines that inform AI-ZTA adoption (CISA,
2023; Ahmad, 2025; Liman Gambo & Almulhem,
2025; Zakhmi et al., 2025)
Studies unrelated to cybersecurity,
or those focusing solely on AI or
ZTA without relevance to
integration or
organizational/remote contexts
Provided empirical data, case studies, systematic
reviews, conceptual models, or
technical/government guidelines relevant to
organizational or remote work environments (Ajish,
2024a; Paul et al., 2024; Nzeako & Shittu, 2024;
Kodi, 2025;
Ajimatanrareje & Agbesi, 2025)
Lacked methodological detail or
relevance to
organizational/remote contexts
Available in full-text format through academic
databases, open-access repositories, or government
publications
(Rose et al., 2020; CISA, 2023)
Inaccessible in full-text format
Clearly described research design, review process,
or technical framework (e.g., PRISMA, case study,
Opinion pieces, editorials, or
sources without
methodological transparency
Page 386
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
survey, architecture model) (Ahmad, 2025; Liman
Gambo & Almulhem, 2025; Zakhmi et al., 2025)
Contributed to understanding AI-ZTA synergy,
predictive analyses, or socio-technical implications
(Karamchand, 2024; Gadkari, 2025; Xu et al., 2025)
Focused solely on traditional
cybersecurity without linkage to
AI-ZTA
Table 3. Eligibility criteria applied in the systematic review of AI-ZTA integration studies (2020–2030),
structured in accordance with PRISMA guidelines.
Information Sources
The literature search was conducted across multiple academic databases, open access repositories, government
and standards, and reference lists to ensure comprehensive coverage of studies addressing AI-ZTA integration.
This produced 25 references spanning foundational frameworks, applied case studies, predictive analyses, and
risk-oriented literature. The inclusion of 25 studies is methodologically sufficient and consistent with PRISMA
standards, balancing breadth and depth while ensuring rigorous appraisal. Comparable systematic reviews in
cybersecurity and Zero-Trust research have also synthesized fewer than 25 studies (e.g., Liman Gambo &
Almulhem, 2025; Mushtaq et al., 2025), demonstrating that a focused pool can yield reliable insights without
sacrificing comprehensiveness. Furthermore, this deliberate restriction aligns with rapid review principles,
where narrowing scope and applying strict inclusion/exclusion criteria enhances efficiency and transparency in
fast-evolving domains such as AI-ZTA.
Figure 2. Distribution of references included in the AI-ZTA systematic review (2020–2030), categorized by
source type.
Source Type
Specific References
Last Searched/Consulted
Academic
Databases
Ajish (2024a, 2024b); Paul et al. (2024); Karamchand (2024);
Nzeako & Shittu (2024); Kodi (2025); Zakhmi et al. (2025);
Ahmad (2025); Liman Gambo & Almulhem (2025);
Ajimatanrareje & Agbesi
(2025)
February–March 2026
Government &
Standards
Rose et al. (2020); CISA (2023); NIST SP 800-207; other
government cybersecurity guidelines
February–March 2026
11
4
5
5
Academic Databases
Government & Standards
Open Access Repositories
Reference Lists
Page 387
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Open Access
Repositories
Gadkari (2025); Ajimatanrareje & Agbesi (2025) [open-
access]; Liman Gambo & Almulhem (2025) [open-access];
institutional archives; MDPI open-access studies
February–March 2026
Reference
Lists
Xu et al. (2025); Ucheji (2026); Rodrigues (2026);
Gartner (2026); citation chasing from bibliographies
February–March 2026
Table 4. Information sources consulted for the AI-ZTA systematic review (2020–2030), categorized into
academic databases, government and standards publications, open-access repositories, and reference lists.
Search Strategy
The search strategy for this systematic review was conducted between February and March 2026 using a
structured approach to ensure comprehensive coverage of literature on AI-ZTA integration. Search terms
included “Artificial Intelligence,” “AI,” “Zero Trust Architecture,” “ZTA,” and “cybersecurity,” with
adjustments for “remote work” and “organizational security” to capture studies in decentralized contexts (Ajish,
2024a; Nzeako & Shittu, 2024). Filters restricted results to publications dated between 2020 and 2030, written
in English, and accessible in full text. The search spanned academic databases such as IEEE Xplore, ACM
Digital Library, MDPI, SpringerLink, and Elsevier, as well as government and standards publications including
NIST SP 800‑207 (Rose et al., 2020) and CISA (2023). Open‑access repositories and institutional archives were
also consulted to ensure inclusivity, while citation tracking identified additional studies through bibliographies
and cross‑references (Ajimatanrareje & Agbesi, 2025). This multi‑layered strategy yielded 25 references
encompassing foundational frameworks, applied case studies, predictive analyses, and risk‑oriented literature.
By integrating diverse sources and employing both database queries and citation chasing, the review captured
the breadth of scholarship relevant to AI-ZTA integration across organizational, cloud, infrastructure, and remote
workforce contexts.
Study Selection
The study selection process began with the identification of 50 records across academic databases, government
publications, open-access repositories, and institutional archives. Titles and abstracts were screened to exclude
studies outside the 2020-2030 timeframe or those lacking relevance to AI-ZTA integration. Full-text assessment
further refined the pool by applying methodological quality checks and contextual relevance. After this rigorous
process, 25 studies were retained, representing foundational frameworks, applied case studies, predictive
analyses, and risk-oriented literature. This progression is illustrated in Figure 3, which summarizes the study
selection process using the PRISMA framework and shows how the initial pool of records was systematically
narrowed to the final 25 studies.
Page 388
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Figure 3. PRISMA 2020 flow diagram showing 50 records identified, with 25 studies meeting eligibility criteria
and included in the final synthesis.
Figure 4 provides a detailed breakdown of the sources consulted. IEEE Xplore contributed eight core articles,
ACM Digital Library five, MDPI six, and Springer/Elsevier seven peer-reviewed studies. Specialized journals
added ten systematic review references, while government repositories such as NIST SP 800-207 and CISA
guidelines supplied five foundational standards. Citation chasing yielded an additional five references, ensuring
coverage of emerging risks and forward-looking analyses. Together, the diagram and table demonstrate a
transparent and comprehensive selection process, ensuring that the final synthesis captures both technical
effectiveness and sociotechnical implications of AI-ZTA integration.
Page 389
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Figure 4. Sources and databases consulted during study identification, totaling 50 records before screening and
25 included in the final synthesis.
Data Collection Procedure
To ensure transparency and comparability across all included studies, the evidence was systematically organized
into a structured table. Table 5 presents the 25 references retained in the final synthesis, detailing their author/s
and year, source type, access status, focus on AI-ZTA integration, and key findings. This process ensured that
all relevant information was captured in a uniform manner, aligned with the eligibility criteria presented in Table
3. By consolidating these characteristics, the table provides a clear overview of the methodological diversity
and thematic contributions that underpin the review’s comparative analysis.
Author(s)
& Year
Source Type
Access
Status
Focus on AI-ZTA
Integration
Key Findings
Rose et al.
(2020)
Government
report (NIST
SP 800-207)
Open access
Foundational framework
Established Zero Trust principles;
baseline for AI integration.
CISA
(2023)
Government
guideline
Open access
Foundational framework
Defined maturity model;
informed AI-ZTA adoption.
Cloud
Security
Tech Ref.
Arch.
(2022)
Government
guideline
Open access
Foundational
framework
Cloud ZTA standards; groundwork for AI
integration.
re
IEEE Xplo
ACM Digital
Library
MDPI
(
Open
-
ss)
Acce
Springer / Elsevier
ResearchGa
te /
Institution
al
Archives
d
Specialize
RG,
Journals (SS
IRJMETS
,
RMT,
WJARR, IJS
IJRSI)
NIST /
Government
Repositories
Citation Searching
Bibliographies
(
)
Records Identified
8
5
6
7
4
10
5
5
8
5
6
7
4
10
5
5
0
2
4
6
8
10
12
Page 390
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
NSTAC
(2022)
Government
report
Open access
Foundational
framework
Linked Zero Trust with trusted identity;
policy recommendations.
Gambo &
Almulhem
(2025)
Preprint
(arXiv)
Open access
Systematic
review
Synthesized ZTA literature; highlighted AI’s
role in anomaly detection.
Liman
Gambo &
Almulhem
(2025)
Journal article
Subscription
Systematic
review
Comprehensive review of ZTA; positioned
AI as resilience enabler.
Campbell
(2026)
Preprint
(Preprints.org)
Open access
Reference
architecture
Proposed assurance framework for AI-ZTA
in organizational contexts.
Ajish (2024a)
Journal article
Subscription
Risk-oriented
Showed AI enhances ZTA in remote work
via anomaly detection.
Ajish (2024b)
Journal article
Open access
Foundational
framework
Analyzed AI’s role in strengthening ZTA
with automation.
Paul et al.
(2024)
Journal article
Open access
Risk-oriented
Proposed synergistic AI-ZTA framework;
resilience against insider threats.
Karamchan d
(2024)
Journal article
Open access
Risk-oriented
Demonstrated AI-ZTA synergy mitigating
advanced threats.
Nzeako &
Shittu
(2024)
Journal article
Open access
Applied case
study
Implemented ZTA in cloud with AI;
improved access control.
Ajimatanrar
eje &
Agbesi
(2025)
Journal article
Open access
Applied case
study
AI-powered ZTA for critical infrastructure;
resilience against attacks.
Ofili et al.
(2025)
Journal article
Open access
Applied case
study
AI-ZTA for federal cloud; compliance with
CISA standards.
Srivastava
(2025)
Journal article
Open access
Applied case
study
Real-time AI-driven threat detection
integrated with ZTA.
Page 391
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Rodrigues
(2026)
Journal article
Open access
Socio-technic
al analysis
Analyzed unified AI-ZTA platforms;
highlighted
workforce/training challenges.
Sastry
(2025)
Journal article
Subscription
Applied case
study
Examined ZTA implementation in
dynamic workforce settings.
Xu et al.
(2025)
Journal article
Open access
Predictive
analysis
Surveyed risks of generative AI eroding
ZTA; suggested safeguards.
Ucheji
(2026)
Journal article
Open access
Predictive
analysis
Tested AI-ZTA in remote workforce;
proactive detection and response.
Zakhmi et al.
(2025)
Journal article
Open access
Systematic
review
Reviewed AI-ZTA in healthcare;
resilience against AI-driven threats.
Cao et al.
(2024)
Journal article
Subscription
Applied
solutions
Explored
automation/orchestration of ZTA; identified
challenges.
Chawande
(2024)
Journal article
Open access
Risk-oriented
Adaptive ZTA with
AI/automation; emphasized dynamic
enforcement.
Chokkanath
an et al.
(2024)
Conference
paper
(CSITSS)
Subscription
Applied case
study
AI-driven ZTA resilience; demonstrated
enhanced cyber defense.
Ueno et al.
(2024)
Conference
paper (CHI)
Subscription
Risk-oriented
Explored trust in human-AI interaction;
implications for AI-ZTA adoption.
Gartner
(2023,
2026)
Analyst reports
Restricted
Predictive
analysis
Forecasted ZTA growth and AI integration;
noted vulnerabilities.
Table 5. Evidence synthesis of 25 studies on AI-enhanced Zero Trust Architecture (2020–2030), categorized by
source type, access status, AI-ZTA integration focus, and key findings, in accordance with PRISMA guidelines.
Data Item
To ensure consistency and comparability across studies, specific outcomes and variables were identified for
extraction. Outcomes Sought focused on the effectiveness of AI-ZTA integration in mitigating cybersecurity
risks, while other variables Sought captured contextual and methodological details necessary for comparative
analysis. The following tables summarize the data items sought.
Outcomes Sought
Outcome Domain
Definition
Examples of Measures
Insider threat mitigation
Reduction of unauthorized insider
activity
Access logs, anomaly detection reports
Data breach prevention
Protection of sensitive
organizational data
Breach frequency, encryption
effectiveness
Advanced cyberattack
resilience
Defense against sophisticated
threats
Ransomware/phishing prevention rates
Page 392
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Automation & policy
enforcement
AI-driven enforcement of
ZTA protocols
Adaptive access control,
automated authentication
Incident response effectiveness
Speed and accuracy of
detecting/responding to attacks
Mean time to detect/respond
Workforce adaptability &
training
Ability of staff to adopt AI-ZTA
practices
Training completion rates, compliance
audits
Governance & compliance
Alignment with regulatory and
organizational standards
Audit trails, adherence to
CISA/NIST guidelines
Predictive risk management
Anticipation of emerging threats
(e.g., generative AI risks)
Forecast models,
safeguard recommendations
Table 6. Outcomes sought in the systematic review of AI-ZTA integration (2020–2030), defining domains of
effectiveness such as insider threat mitigation, data breach prevention, resilience, automation, incident response,
workforce adaptability, governance, and predictive risk management.
Other Variables Sought
Variable
Definition
Assumptions for Missing/Unclear Data
Study characteristics
Author(s), year, study design
Categorized based on explicit description;
inferred from publication type if unclear
Contextual setting
Organizational, remote workforce,
cloud, or
infrastructure
Defaulted to “organizational security” unless
clearly tied to cloud/infrastructure
Intervention details
AI techniques and ZTA
components used
If unspecified, assumed anomaly detection and
automation as common practices
Industry relevance
Sector or environment of
application
Inferred from study context; if absent, treated
as general organizational security
Funding sources
Institutional or external support
Recorded only when explicitly mentioned;
otherwise treated as independent
Reported limitations
Methodological constraints or risks
noted
Extracted when available; if absent, assumed
not reported
Access status
Open access, subscription, or restricted
Verified through publication metadata
Integration focus
Framework, case study, risk analysis,
predictive analysis
Assigned based on primary contribution
Table 7. Contextual and methodological variables sought in the systematic review of AI-ZTA integration (2020–
2030), including study characteristics, settings, interventions, industry relevance, funding, limitations, access
status, and integration focus The outcomes and variables summarized in Tables 6 and 7 provided the structured
basis for evidence extraction. Outcomes defined measurable domains of AI-ZTA effectiveness including insider
threat mitigation, data breach prevention, resilience, automation, incident response, workforce adaptability,
governance, and predictive risk management while variables captured methodological and contextual details
across diverse study designs.
This framework ensured that all 25 included studies were evaluated on comparable grounds. It also enabled the
synthesis to trace the chronological progression of AI-ZTA adoption, from foundational frameworks to
predictive analyses, and to group studies thematically into categories such as frameworks, risk-oriented
literature, applied case studies, systematic reviews, predictive analyses, and socio-technical perspectives. By
Page 393
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
structuring the evidence in this way, the review supports a rigorous comparative synthesis and highlights both
the technical effectiveness and practical adaptability of AI-ZTA in organizational and remote work contexts.
Limitations
While this systematic review was conducted in accordance with the PRISMA framework to ensure transparency
and replicability, several methodological constraints should be acknowledged:
• Source Accessibility: Only studies available in full-text format through academic databases, government
repositories, or open-access archives were included. This may have excluded potentially relevant
research that was inaccessible due to subscription restrictions or proprietary limitations.
• Timeframe Restriction (2020–2030): The review deliberately limited its scope to publications within this
decade to capture contemporary developments in AI-ZTA integration. While this enhances relevance, it
excludes earlier foundational work on AI or Zero-Trust principles that may have provided historical
context.
• Language Bias: The search strategy restricted results to English-language publications. This introduces
a potential bias by excluding non-English studies that may contain valuable insights, particularly from
regions with different cybersecurity adoption trajectories.
• Database Coverage: Although multiple databases (IEEE Xplore, ACM Digital Library, MDPI,
SpringerLink, Elsevier) and government sources (NIST, CISA) were consulted, the review may not have
captured all relevant studies, especially those published in niche or regional outlets.
• Methodological Transparency of Sources: Studies lacking clear methodological detail were excluded to
maintain rigor. While this strengthens reliability, it may have limited the diversity of perspectives,
particularly from industry reports or practitioner-oriented publications.
• Citation Chasing Dependence: Some predictive and socio-technical perspectives were identified through
citation chasing rather than direct database queries. This approach, while valuable, may introduce
selection bias by favoring studies referenced in already included literature.
• Review Duration: The review was conducted within a 90-day timeframe. While this period allowed for
a structured and systematic search, it inherently limited the inclusion of studies published after the cutoff
date, reflecting the balance between timeliness and comprehensiveness in fast-evolving domains such as
AI-ZTA.
These limitations highlight the importance of cautious interpretation. The findings provide a robust synthesis of
AI-ZTA integration within remote and organizational contexts, but they should be understood as representative
rather than exhaustive. Future reviews could expand coverage by incorporating multilingual sources, extending
the timeframe, and including grey literature to capture a broader spectrum of evidence.
Synthesis
To make the findings clearer and easier to follow, the synthesis not only explains how AI-ZTA adoption has
progressed over time but also includes a visual summary. Figure 5 shows the journey from early frameworks to
predictive analyses, giving both the detailed evidence and the bigger picture at a glance.
Across the 25 studies reviewed, the evidence highlights how AI-ZTA has proven both technically effective and
practically adaptable in organizational, cloud, and remote workforce settings. Grouping the evidence
thematically reveals a clear chronological and conceptual progression:
• Foundational Frameworks (2020–2023) Government guidelines (Rose et al., 2020; CISA, 2023) and early
conceptual models established the baseline principles of Zero-Trust and positioned AI as a potential
Page 394
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
enabler of anomaly detection and automation. These works provided the methodological and technical
foundation for subsequent applied studies.
• Risk-Oriented Literature (2024) Studies such as Ajish (2024a), Paul et al. (2024), and Karamchand (2024)
explored the vulnerabilities of decentralized workforces and emphasized AI’s role in mitigating insider
threats and advanced persistent attacks. This stage highlighted both opportunities and risks, including
ethical concerns and the erosion of trust through generative AI.
• Applied Case Studies (2024–2025) Case studies (Nzeako & Shittu, 2024; Ajimatanrareje & Agbesi, 2025;
Ofili et al., 2025) demonstrated practical deployments of AI-ZTA in cloud and critical infrastructure
environments. These implementations validated the framework’s scalability, improved access control, and
compliance with government standards, while also identifying workforce training and cost challenges.
• Predictive Analyses (2025–2026) Forward-looking studies (Xu et al., 2025; Ucheji, 2026; Gartner, 2026)
forecasted emerging risks, particularly the impact of generative AI on Zero-Trust principles, and proposed
safeguards to preserve resilience. These analyses underscored the need for continuous evaluation and
adaptive governance.
Mapping these categories against the outcomes defined in Table 6 confirms systematic alignment: insider threat
mitigation, data breach prevention, resilience against advanced cyberattacks, automation and policy
enforcement, incident response effectiveness, workforce adaptability, governance and compliance, and
predictive risk management. Collectively, the evidence positions AI-ZTA as a cornerstone of future
cybersecurity strategies, while emphasizing that its long-term success depends on robust safeguards, workforce
training, regulatory compliance, and iterative evaluation mechanisms.
Figure 5. Chronological and thematic synthesis of AI-ZTA integration studies (2020–2030), illustrating the
progression from foundational frameworks (2020–2023) to risk‑oriented literature
(2024), applied case studies (2024–2025), and predictive analyses (2025–2026). This visual summary highlights
both the technical effectiveness and the persistent limitations—such as cost, workforce readiness, and ethical
concerns—that continue to shape AI-ZTA adoption.
Page 395
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
RESULTS
The synthesis of the 25 included studies provided a comprehensive view of how AI-ZTA has evolved between
2020 and 2030. Evidence was organized into thematic categories that reflect both technical effectiveness and
sociotechnical implications, allowing the review to trace the chronological progression from foundational
frameworks to predictive analyses. This structured approach highlights not only the diversity of contexts in
which AI-ZTA has been applied but also the recurring challenges and opportunities that shape its adoption.
To present these findings in a clear and systematic manner, the results are discussed across four domains aligned
with the research questions and objectives. These domains include the applications of AI-ZTA across
organizational, cloud, and remote workforce contexts; the challenges and limitations encountered in adoption,
the effectiveness of AI-ZTA in mitigating cybersecurity risks, and the future opportunities and risks forecasted
for its continued integration. Each subsection builds upon the evidence base, illustrating both the technical
contributions and the broader implications for organizational resilience and cybersecurity governance.
Applications of AI-ZTA Across Contexts (RQ1 / RO1)
The review found that AI-ZTA has been applied in many settings—inside organizations, in cloud systems, and
across remote workforces. Early government frameworks (Rose et al., 2020; CISA, 2023; Cloud Security Tech
Ref. Arch., 2022; NSTAC, 2022) laid the groundwork by defining principles like continuous authentication and
identity verification. Later, applied case studies showed how these ideas worked in practice. For example,
Nzeako & Shittu (2024), Ajimatanrareje & Agbesi (2025), and Ofili et al. (2025) demonstrated AI-ZTA in cloud
and critical infrastructure, improving access control, compliance, and resilience. Srivastava (2025) and Sastry
(2025) extended these applications to dynamic workforce environments, proving adaptability in remote and
hybrid contexts.
To make this progression clearer, Figure 6 shows a timeline of how AI-ZTA adoption evolved— from
foundational standards (2020–2023), to applied deployments in cloud and infrastructure (2024–2025), and
finally to dynamic adaptation for remote and hybrid work (2025–2026). This visual highlights how AI-ZTA
steadily expanded from theory into practice across different organizational contexts.
Figure 6. Applications of AI-ZTA Across Contexts (RQ1/RO1)
Challenges and Limitations in Adoption (RQ2 / RO2)
Several studies pointed out barriers to adopting AI-ZTA. Rodrigues (2026) stressed the lack of workforce
training and governance issues, while Xu (2025) warned that generative AI could weaken Zero Trust principles.
Gartner (2026) added concerns about surveillance, bias, and vulnerabilities in predictive models. Case studies
like Ajimatanrareje & Agbesi (2025) and Sastry (2025) also showed that high costs and integration complexity
are common problems. These findings suggest that while AI-ZTA is technically strong, its success depends on
organizations being ready, compliant, and guided by ethical safeguards.
Page 396
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
To make these challenges clearer, Figure 7 shows a bar chart of the most common issues—training gaps, ethical
and governance concerns, risks from generative AI, and high costs. The visual highlights how often these
problems appear in the literature, reminding us that sustainability of AI-ZTA is not just about technology but
also about people, policies, and resources.
Figure 7. Challenges and Limitations in Adoption (RQ2/RO2)
Effectiveness in Mitigating Cybersecurity Risks (RQ3 / RO3)
The evidence clearly shows that AI-ZTA makes organizations more resilient against cyber threats. Studies
highlighted how AI helps with anomaly detection, insider threat prevention, and adaptive enforcement (Ajish,
2024a; Paul et al., 2024; Karamchand, 2024; Chawande, 2024). Case studies confirmed fewer unauthorized
access incidents, faster breach detection, and stronger compliance with government standards. Systematic
reviews (Gambo & Almulhem, 2025; Liman Gambo & Almulhem, 2025; Zakhmi et al., 2025) reinforced these
findings across healthcare, cloud, and organizational contexts. Together, the studies show that AI-ZTA is
effective in stopping insider threats, preventing data breaches, and defending against advanced attacks.
To make this comparison clearer, Figure 8 uses a radar chart to show how AI-ZTA performs against traditional
ZTA across five areas: insider threat mitigation, data breach prevention, advanced attack defense, compliance,
and incident response speed. The chart highlights AI-ZTA’s consistent edge, showing how automation and
proactive defense make it stronger and more adaptable than traditional approaches.
Figure 8. Effectiveness in Mitigating Cybersecurity Risks (RQ3/RO3)
Page 397
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Future Opportunities and Risks (RQ4 / RO4 & RO5)
Studies forecast both opportunities and risks for AI-ZTA adoption. On the positive side, researchers highlight
proactive detection, automated policy enforcement, and assurance frameworks that can strengthen
organizational resilience (Xu, 2025; Ucheji, 2026; Campbell, 2026). At the same time, risks remain. Generative
AI may erode Zero Trust principles, while ethical concerns about surveillance, bias, and governance gaps
continue to surface (Gartner, 2023, 2026; Rodrigues, 2026). Workforce adaptability also emerges as a challenge,
showing that technology alone is not enough. Sustaining effectiveness will require ongoing evaluation,
regulatory alignment, and training.
To capture this balance, Figure 9 presents a split timeline. The upper branch highlights opportunities such as
proactive detection, automated enforcement, and assurance frameworks, while the lower branch shows risks like
generative AI erosion, ethical concerns, and workforce challenges. This visual makes clear that the future of AI-
ZTA depends not only on innovation but also on responsible governance and human readiness.
Figure 9. Future Opportunities and Risks (RQ4/RO4 & RO5)
DISCUSSIONS
The findings of this systematic review highlight that AI-ZTA consistently strengthens organizational resilience
against insider threats, data breaches, and advanced cyberattacks. Foundational frameworks such as Rose et al.
(2020) and CISA (2023) laid the groundwork by defining the principle of “never trust, always verify,” while
applied case studies (Nzeako & Shittu, 2024; Ajimatanrareje & Agbesi, 2025) demonstrated how AI integration
enhances access control and incident response in real‑world deployments. Predictive analyses (Xu et al., 2025;
Ucheji, 2026) further emphasized both opportunities and risks, particularly the erosion of Zero Trust principles
through generative AI. Collectively, these studies illustrate a clear trajectory from conceptual frameworks to
applied solutions and risk‑oriented literature.
A recurring theme across the evidence is the shift from reactive defense to proactive resilience. AI’s role in
anomaly detection, automated policy enforcement, and rapid incident response transforms ZTA from a static
model into a dynamic, adaptive system. This proactive stance enables organizations to anticipate and neutralize
threats before they escalate, reducing mean time to respond (MTTR) and improving overall resilience. However,
sociotechnical challenges remain significant. Rodrigues (2026) highlighted workforce training gaps and high
implementation costs, while Gartner (2026) raised ethical concerns regarding surveillance and algorithmic bias.
These findings suggest that while AI-ZTA is technically effective, its sustainability depends on governance,
compliance, and human adaptability.
Page 398
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
Limitations of Reviewed Studies
Beyond the practical challenges of adoption, the reviewed studies themselves reveal important limitations.
Several case studies were conducted in controlled or pilot environments, which may not fully capture the
complexity of large‑scale organizational deployments (Ajimatanrareje & Agbesi, 2025). Predictive analyses
often relied on theoretical models rather than empirical validation, raising questions about their long‑term
applicability (Xu et al., 2025). Furthermore, sociotechnical concerns such as workforce readiness and ethical
implications were acknowledged but rarely supported by quantitative data (Rodrigues, 2026; Gartner, 2026).
These gaps limit the generalizability of findings and highlight the need for longitudinal, cross‑industry research
to confirm the effectiveness and sustainability of AI-ZTA integration.
Visual Summary
Chronological analysis reveals a maturation trajectory in AI‑ZTA adoption. Early studies (2020– 2023) focused
on foundational frameworks, mid‑decade research (2024–2025) emphasized applied deployments and
risk‑oriented literature, and later analyses (2025–2026) forecasted future risks and safeguards. Figure 5 provides
a visual timeline of this progression, reinforcing that while technical effectiveness has been demonstrated,
unresolved limitations persist across stages, underscoring the importance of continuous evaluation.
CONCLUSIONS
This systematic review confirms that AI-ZTA integration is a critical paradigm for securing data in decentralized
work environments. By combining Zero-Trust principles with AI-driven automation, organizations can
effectively mitigate insider threats, prevent breaches, and enhance resilience against sophisticated cyberattacks.
The evidence demonstrates that AI-ZTA is not only technically effective but also practically adaptable across
industries, including cloud infrastructure, healthcare, and government systems.
Despite these strengths, unresolved challenges remain. High implementation costs, workforce training gaps,
ethical concerns, and regulatory compliance issues pose barriers to widespread adoption. Studies such as
Rodrigues (2026) and Gartner (2026) emphasize that without addressing these sociotechnical dimensions, AI-
ZTA risks becoming a technically sound but practically limited solution. Therefore, while AI-ZTA offers
transformative potential, its long-term success depends on balancing technical innovation with organizational
readiness and ethical safeguards. AI-ZTA represents both an opportunity and a challenge. Its effectiveness hinges
on robust safeguards, continuous monitoring, and alignment with governance structures. As organizations
increasingly adopt remote and hybrid work models, AI-ZTA provides a scalable and adaptable framework for
cybersecurity resilience. However, its sustainability requires not only technical refinement but also cultural and
regulatory alignment to ensure trust, accountability, and ethical use of AI in security contexts.
RECOMMENDATIONS
To ensure the successful adoption of AI-ZTA, governments and industry regulators should establish clear
frameworks aligned with established standards such as NIST SP 800-207 and CISA guidelines. These
frameworks provide a consistent baseline for organizations, reducing ambiguity and ensuring that practices are
uniformly applied across industries. Continuous compliance audits and transparent reporting mechanisms
should also be mandated to strengthen accountability and provide measurable benchmarks for evaluating
effectiveness. Anticipating emerging risks, particularly those posed by generative AI, is essential; safeguards
must be embedded into policy frameworks to protect the integrity of Zero-Trust principles and maintain
resilience against evolving threats.
Organizations themselves must prioritize workforce training to bridge skill gaps in AI-ZTA implementation.
Employees are central to the success of any cybersecurity framework, and structured training programs should
emphasize both technical competencies and ethical considerations. Phased deployment strategies are equally
important, as they allow organizations to adopt AI-ZTA gradually, minimizing disruption and spreading costs
over time. This iterative approach ensures that systems are refined before scaling, while fostering collaboration
Page 399
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
across IT, compliance, and leadership teams to address both technical and sociotechnical challenges inherent in
cybersecurity transformation.
From a technical perspective, AI-driven anomaly detection and automated policy enforcement should be
integrated as baseline components of Zero-Trust systems. These features transform the framework from a static
model into a dynamic, adaptive defense capable of responding to evolving threats. At the same time,
organizations must develop safeguards against generative AI risks, which can undermine authentication and
identity verification processes. Countermeasures such as adversarial testing, continuous monitoring, and AI-
driven identity verification are critical to maintaining trust boundaries and ensuring resilience in decentralized
environments.
Further research is needed to evaluate AI-ZTA’s long-term effectiveness beyond initial deployment.
Longitudinal studies can provide insights into sustainability, scalability, and adaptability across different
organizational contexts. Sector-specific applications in healthcare, education, and government should also be
explored to identify tailored strategies that address unique risks. Finally, the ethical implications of AI-enabled
surveillance must be investigated, with frameworks developed to ensure responsible use. By combining
technical safeguards with ongoing research, AI-ZTA can evolve into a sustainable and trusted cybersecurity
paradigm capable of adapting to future challenges.
REFERENCES
1. Ajimatanrareje, G. A., & Agbesi, J. S. (2025c). AI-Powered Zero Trust Architectures for Critical
Infrastructure Protection: A Comprehensive framework for Next-Generation CyberSecurity. International
Journal of Scientific Research and Modern Technology., 40–56. https://doi.org/10.38124/ijsrmt.v4i9.792
2. Ajish, D. (2024b). The significance of artificial intelligence in zero trust technologies: a comprehensive
review. Journal of Electrical Systems and Information Technology, 11(1).
https://doi.org/10.1186/s43067-024-00155-z
3. Campbell, R. (2026). Zero Trust for AI Systems: A reference architecture and assurance framework.
Preprints.org. https://doi.org/10.20944/preprints202602.0085.v1
4. Cao, Y., Pokhrel, S. R., Zhu, Y., Doss, R., & Li, G. (2024). Automation and Orchestration of Zero Trust
architecture: Potential solutions and challenges. Machine Intelligence Research, 21(2), 294–317.
https://doi.org/10.1007/s11633-023-1456-2
5. Chawande, S. (2024). Adaptive zero trust with AI and Automation. World Journal of Advanced
Engineering Technology and Sciences, 13(2), 751–763.
https://doi.org/10.30574/wjaets.2024.13.2.0589
6. Chokkanathan, K., Karpagavalli, S., Priyanka, G., Vanitha, K., Anitha, K., & Shenbagavalli, P. (2024).
AI-Driven Zero Trust Architecture: Enhancing Cyber-Security resilience. Proceedings of CSITSS
2024, 1–6. https://doi.org/10.1109/csitss64042.2024.10816746
7. CISA. (2023). Zero Trust Maturity Model. https://www.cisa.gov/sites/default/files/2023-
04/zero_trust_maturity_model_v2_508.pdf
8. Cloud Security Technical Reference Architecture. (2022). Federal Risk and Authorization Management
Program. Cybersecurity and Infrastructure Security Agency.
https://www.cisa.gov/sites/default/files/publications/Cloud%20Security%20Technica
l%20Reference%20Architecture.pdf Top Strategic Technology Trends for 2026
9. Gambo, M. L., & Almulhem, A. (2025). Zero Trust Architecture: A Systematic Literature Review. arXiv
(Cornell University). https://doi.org/10.48550/arxiv.2503.11659
10. Gartner. (2026). Zero trust 2030 – Business operating system. Gartner Research.
https://www.gartner.com/en/articles/top-technology-trends-2026
11. Gartner. (2023, January 27). Gartner predicts growth in zero-trust programs by 2026, despite unprotected
areas vulnerabilities. Business of Tech. https://businessof.tech/2023/01/27/gartner-
predicts-growth-in-zero-trust-programsby-2026-despite-unprotected-areas-vulnerabilities
12. Karamchand, G. (2024). Zero trust and AI: A synergistic approach to next-generation cyber threat
mitigation. World Journal of Advanced Research and Reviews, 24(3), 3374–3387.
https://doi.org/10.30574/wjarr.2024.24.3.3883
Page 400
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue IV, April 2026
13. Liman Gambo, M., & Almulhem, A. (2025). Zero trust architecture: A systematic literature review.
Journal of Network and Systems Management.
https://doi.org/10.1007/s10922-025-09998-x
14. Nizamuddin, M. Investigating the cybersecurity risks of remote work: a systematic literature review of
organizational vulnerabilities and mitigation strategies. Int. J. Inf. Secur. 24, 187 (2025).
https://doi.org/10.1007/s10207-025-01095-z
15. NSTAC. (2022). Zero trust and trusted identity management: Report to the President. National Security
Telecommunications Advisory Committee.
16. https://www.cisa.gov/sites/default/files/publications/NSTAC%20Report%20to%20th
e%20President%20on%20Zero%20Trust%20and%20Trusted%20Identity%20Manag ement.pdf
17. Nzeako, C., & Shittu, A. (2024). Implementing zero trust security models in cloud computing
environments. World Journal of Advanced Research and Reviews, 24(3), 1647–1660.
https://doi.org/10.30574/wjarr.2024.24.3.3500
18. Ofili, B. T., Erhabor, E. O., & Obasuyi, T. (2025). Enhancing federal cloud security with AI: Zero trust,
threat intelligence and CISA compliance. World Journal of Advanced Research and Reviews, 25(2),
2377–2400. https://doi.org/10.30574/wjarr.2025.25.2.0620
19. Paul, J., Kessie, J. D., & Salawudeen, M. D. (2024). Zero trust architecture and AI: A synergistic
approach to next-generation cybersecurity frameworks. International Journal of Science and Research
Archive, 13(2), 4159–4169.
https://doi.org/10.30574/ijsra.2024.13.2.2583
20. Rodrigues, A. (2026). AI-integrated zero trust architectures: A socio-technical analysis of unified
enterprise cybersecurity platforms. International Journal for Research in Applied Science and
Engineering Technology, 14(1), 946–953.
https://doi.org/10.22214/ijraset.2026.76981
21. Rose, S., Borchert, O., Mitchell, S., & Connelly, S. (2020d). Zero trust architecture.
https://doi.org/10.6028/nist.sp.800-207
22. Sabin, J. (2021). The future of security in a remote-work environment. Network Security, 2021(10), 15–
17. https://doi.org/10.1016/s1353-4858(21)00118-5
23. Sastry, R. (2025). Zero trust architecture implementation in dynamic workforce environments.
International Journal of Computer Engineering and Technology, 16(1), 92–99.
https://doi.org/10.34218/IJCET_16_01_092
24. Srivastava, S. (2025). Real time AI-driven threat detection with the integration of zero trust security
framework. TEJAS Journal of Technologies and Humanitarian Science, 4(4), 25–37.
https://doi.org/10.63920/tjths.44004
25. Ucheji, C. (2026). The future of zero-trust security architecture with AI automation. International Journal
of Research and Scientific Innovation, 13(1), 700–713.
https://doi.org/10.51244/ijrsi.2026.13010060
26. Ueno, H., Sawa, Y., Kim, J., Urakami, T., Oura, K., & Seaborn, K. (2024). Trust in human-AI interaction:
Scoping out models, measures, and methods. Proceedings of the CHI Conference on Human Factors in
Computing Systems Extended Abstracts, 1–7. https://doi.org/10.1145/3491101.3519772
27. Xu, J., Gondal, I., Yi, X., Susnjak, T., Watters, P., & McIntosh, S. (2025). The erosion of cybersecurity
zero-trust principles through generative AI: A survey on the challenges and future directions. Journal of
Cybersecurity and Privacy, 5(4), 87. https://doi.org/10.3390/jcp5040087
28. Zakhmi, K., Ushmani, A., Mohanty, M. R., Agrawal, S., Banduni, A., & Rao, S. S. K. (2025). Evolving
zero trust architectures for AI-driven cyber threats in healthcare and other high-risk data
environments: A systematic review. Cureus.
https://doi.org/10.7759/cureus.85446