Recent advances in digital infrastructures, cloud services, blockchain platforms, Internet of Things (IoT), and

artificial intelligence (AI) have significantly increased the complexity of cybersecurity threats while intensifying

concerns related to data privacy and system trust.

In response, contemporary research has focused on integrating intelligent data-driven techniques with secure and

privacy-aware mechanisms to counter sophisticated cyberattacks.

This review provides a structured foundation to support future research on intelligent and trustworthy

cybersecurity systems.

Keywords: Cybersecurity, Artificial Intelligence, Blockchain Security, Intrusion Detection, Explainable AI

The rapid growth of data-centric technologies has reshaped modern cybersecurity landscapes. Systems deployed

in financial networks, cloud infrastructures, and distributed digital platforms generate large volumes of sensitive

data, making them attractive targets for cyberattacks [1], [2].

Traditional rule-based and perimeter-oriented security mechanisms are increasingly insufficient against

sophisticated threats such as advanced persistent attacks and data-driven exploitation strategies [3], [5].

Consequently, recent research has shifted toward intelligent cybersecurity solutions that leverage machine

Artificial intelligence has become a cornerstone of modern intrusion detection systems. Ghosh et al. [1] proposed

an AI-driven financial cybersecurity framework combining recurrent neural networks (RNNs) with scaled gated

recurrent units (SGRUs) and explainable AI techniques.

Their approach demonstrates strong capability in modeling sequential transaction behavior while offering

interpretability for security analysts. However, the framework introduces increased computational overhead,

raising concerns regarding scalability and real-time deployment.

Other studies within the selected literature explore supervised machine learning techniques for cyberattack

detection in financial and distributed environments [2], [7].

These approaches demonstrate improved detection accuracy compared to traditional methods but rely heavily

on labeled datasets, which limits adaptability to evolving and previously unseen attacks [3], [5].

complexity.

Despite these advantages, existing blockchain-based security solutions face challenges related to scalability,

latency, and system integration [4], [6], [9].

Several studies emphasize that blockchain alone does not provide intelligent threat detection and must be

combined with AI-driven analytics to address complex cyberattack

Explainability has emerged as a critical requirement for intelligent cybersecurity systems, particularly in

financial and regulatory-sensitive domains.

The work by Ghosh et al. [1] demonstrates the benefits of explainable AI in improving analyst trust and decision

transparency. Similarly, conceptual studies highlight the importance of governance, accountability, and system

outperform traditional rule-based approaches in identifying complex attack patterns [1], [7].

Deep learning models are effective in capturing sequential and high-dimensional behaviours, whereas classical

machine learning techniques offer advantages in interpretability and computational efficiency [2].

Blockchain-enabled frameworks enhance trust, auditability, and data integrity but introduce latency and

scalability challenges that limit real-time applicability [4], [6].

Furthermore, most studies evaluate system performance using accuracy-centric metrics, with limited attention

to deployment constraints and system-level efficiency [5], [9].

Table I presents a comprehensive comparative analysis of these approaches and serves as a reference for

identifying opportunities for future research.

Table I. Comparative Analysis of Existing Intelligent and Privacy-Aware Cybersecurity Approaches

Although AI-driven and blockchain-enabled cybersecurity solutions significantly improve detection accuracy

and system trust, existing approaches largely emphasize performance while overlooking scalability,

explainability, and deployment feasibility as unified objectives [1], [6], [7]. Deep learning–based models

demonstrate strong capability in capturing complex attack behaviours but incur high computational overhead

and system complexity [1], [8].

Conversely, blockchain-assisted security frameworks enhance auditability and integrity but struggle with

latency, scalability, and intelligent attack detection when deployed in real-time environments [4], [6], [9].

Furthermore, current evaluation practices focus predominantly on accuracy metrics, limiting insight into system

robustness and operational effectiveness [5], [7].

By addressing these challenges, future research can bridge the gap between conceptual security models and

1. To design an intelligent cybersecurity framework that integrates AI-driven detection with blockchain-

2. To improve system interpretability through explainable AI techniques suitable for security analysts [1],

3. To enhance scalability and efficiency while maintaining high detection accuracy [2], [8].

4. To evaluate system performance across heterogeneous datasets and attack scenarios [7], [9].

1. A unified intelligent cybersecurity architecture combining AI-driven detection and blockchain-enabled

4. S. B. Ehsan, et al., “Blockchain-Based Cybersecurity Solutions for Secure Transactions,” International