Mohsen Hatami

The increasing integration of AI agents into cyber-physical systems (CPS) introduces new security risks that extend beyond traditional cyber or physical threat models. Recent advances in generative AI enable deepfake and semantic manipulation attacks that can compromise agent perception, reasoning, and interaction with the physical environment, while emerging protocols such as the Model Context Protocol (MCP) further expand the attack surface through dynamic tool use and cross-domain context sharing. This survey provides a comprehensive review of security threats targeting AI agents in CPS, with a particular focus on environmental interactions, deepfake-driven attacks, and MCP-mediated vulnerabilities. We organize the literature using the SENTINEL framework, a lifecycle-aware methodology that integrates threat characterization, feasibility analysis under CPS constraints, defense selection, and continuous validation. Through an end-to-end case study grounded in a real-world smart grid deployment, we quantitatively illustrate how timing, noise, and false-positive costs constrain deployable defenses, and why detection mechanisms alone are insufficient as decision authorities in safety-critical CPS. The survey highlights the role of provenance- and physics-grounded trust mechanisms and defense-in-depth architectures, and outlines open challenges toward trustworthy AI-enabled CPS.

Most adaptive learning systems personalize around content sequencing and difficulty adjustment rather than transforming instructional material within the lesson itself. This paper presents the STREAM (Semantic Transformation and Real-Time Educational Adaptation Multimodal) framework. This modular pipeline decomposes multimodal educational content into semantically tagged, pedagogically annotated units for regeneration into alternative formats while preserving source traceability. STREAM is designed to integrate automatic speech recognition, transformer-based natural language processing, and planned computer vision components to extract instructional elements from teacher explanations, slides, and embedded media. Each unit receives metadata, including time codes, instructional type, cognitive demand, and prerequisite concepts, designed to enable format-specific regeneration with explicit provenance links. For a predefined visual-learner profile, the system generates annotated path diagrams, two-panel instructional guides, and entity pictograms with complete back-link coverage. Ablation studies confirm that individual components contribute measurably to output completeness without compromising traceability. This paper reports results from a tightly scoped feasibility pilot that processes a single five-minute elementary STEM video offline under clean audio–visual conditions. We position the pilot’s limitations as testable hypotheses that require validation across diverse content domains, authentic deployments with ambient noise and bandwidth constraints, multiple learner profiles, including multilingual students and learners with disabilities, and controlled comprehension studies. The contribution is a transparent technical demonstration of feasibility and a methodological scaffold for investigating whether within-lesson content transformation can support personalized learning at scale.

Most adaptive learning systems personalize around content sequencing and difficulty adjustment rather than transforming instructional material within the lesson itself. This paper presents the STREAM (Semantic Transformation and Real-Time Educational Adaptation Multimodal) framework. This modular pipeline decomposes multimodal educational content into semantically tagged, pedagogically annotated units for regeneration into alternative formats while preserving source traceability. STREAM integrates automatic speech recognition, transformer-based natural language processing, and planned computer vision components to extract instructional elements from teacher explanations, slides, and embedded media. Each unit receives metadata including timecodes, instructional type, cognitive demand, and prerequisite concepts, enabling format-specific regeneration with explicit provenance links. We report results from a tightly scoped feasibility pilot processing a single five-minute elementary STEM video offline under clean audio-visual conditions. For a predefined visual-learner profile, the system generates annotated path diagrams, two-panel instructional guides, and entity pictograms with complete back-link coverage. Ablation studies confirm individual components contribute measurably to output completeness without compromising traceability. This narrow scope precludes claims about classroom effectiveness, real-time streaming, scalability across content types, or educational impact across learner populations. We position these as testable hypotheses requiring validation across diverse content domains, authentic deployments with ambient noise and bandwidth constraints, multiple learner profiles including multilingual students and learners with disabilities, and controlled comprehension studies. The contribution is a transparent technical demonstration and methodological scaffold for investigating whether within-lesson content transformation can meaningfully support personalized learning at scale.

This study examined using the Metaverse as an educational tool to bridge academic learning and industry readiness in a computer design course. A paired pretest-posttest design was used to assess changes in familiarity, comfort, motivation, and perceived industry relevance among 28 senior undergraduates following a brief Metaverse-based intervention. Statistical analysis revealed significant gains in students’ confidence and readiness, with moderate to large effect sizes. While technical and accessibility challenges were noted, findings showed that even brief, simulation-based Metaverse experiences can positively impact students’ perceptions of emerging technologies. Mixed views on long-term value emphasize the need for infrastructure, faculty training, and hybrid models. This study contributes to immersive education by demonstrating the practical benefits of short-term Metaverse integration in supporting professional skill development.

The rapid expansion of the Metaverse presents complex security challenges, particularly in verifying virtual objects and avatars within immersive environments. Conventional authentication methods, such as passwords and biometrics, often prove inadequate in these dynamic environments, especially as essential infrastructures, such as smart grids, integrate with virtual platforms. Cybersecurity threats intensify as advanced attacks introduce fraudulent data, compromising system reliability and safety. Using the Electric Network Frequency (ENF), a naturally varying signal emitted from power grids, provides an innovative environmental fingerprint to authenticate digital twins and Metaverse entities in the smart grid. This paper provides a comprehensive survey of the ENF as an environmental fingerprint for enhancing Metaverse security, reviewing its characteristics, sensing methods, limitations, and applications in threat modeling and the CIA triad (Confidentiality, Integrity, and Availability), and presents a real-world case study to demonstrate its effectiveness in practical settings. By capturing the ENF as having a unique signature that is timestamped, this method strengthens security by directly correlating physical grid behavior and virtual interactions, effectively combating threats such as deepfake manipulations. Building upon recent developments in signal processing, this strategy reinforces the integrity of digital environments, delivering robust protection against evolving cyber–physical risks and facilitating secure, scalable virtual infrastructures.

Integrating digital twins (DTs) into smart grid systems within the Internet of Smart Grid Things (IoSGT) ecosystem brings novel opportunities but also security challenges. Specifically, advanced machine learning (ML)-based Deepfake technologies enable adversaries to create highly realistic yet fraudulent DTs, threatening critical infrastructures’ reliability, safety, and integrity. In this paper, we introduce Authenticating Networked Computerized Handling of Representations for Smart Grid security (ANCHOR-Grid), an innovative authentication framework that leverages Electric Network Frequency (ENF) signals as real-world anchors to secure smart grid DTs at the frontier against Deepfake attacks. By capturing distinctive ENF variations from physical grid components and embedding these environmental fingerprints into their digital counterparts, ANCHOR-Grid provides a robust mechanism to ensure the authenticity and trustworthiness of virtual representations. We conducted comprehensive simulations and experiments within a virtual smart grid environment to evaluate ANCHOR-Grid. We crafted both authentic and Deepfake DTs of grid components, with the latter attempting to mimic legitimate behavior but lacking correct ENF signatures. Our results show that ANCHOR-Grid effectively differentiates between authentic and fraudulent DTs, demonstrating its potential as a reliable security layer for smart grid systems operating in the IoSGT ecosystem. In our virtual smart grid simulations, ANCHOR-Grid achieved a detection rate of 99.8% with only 0.2% false positives for Deepfake DTs at a sparse attack rate (1 forged packet per 500 legitimate packets). At a higher attack frequency (1 forged packet per 50 legitimate packets), it maintained a robust 97.5% detection rate with 1.5% false positives. Against replay attacks, it detected 94% of 5 s-old signatures and 98.5% of 120 s-old signatures. Even with 5% injected noise, detection remained at 96.5% (dropping to 88% at 20% noise), and under network latencies from <5 ms to 200 ms, accuracy ranged from 99.9% down to 95%. These results demonstrate ANCHOR-Grid’s high reliability and practical viability for securing smart grid DTs. These findings highlight the importance of integrating real-world environmental data into authentication processes for critical infrastructure and lay the foundation for future research on leveraging physical world cues to secure digital ecosystems.

Modern education faces persistent challenges, including disengagement, inequitable access to learning resources, and the lack of personalized instruction, particularly in virtual environments. In this perspective, we envision a transformative Metaverse classroom model, the Multi-layered Immersive Learning Environment (Meta-MILE) to address these critical issues. The Meta-MILE framework integrates essential components such as immersive infrastructure, personalized interactions, social collaboration, and advanced assessment techniques to enhance student engagement and inclusivity. By leveraging three-dimensional (3D) virtual environments, artificial intelligence (AI)-driven personalization, gamified learning pathways, and scenario-based evaluations, the Meta-MILE model offers tailored learning experiences that traditional virtual classrooms often struggle to achieve. Acknowledging potential challenges such as accessibility, infrastructure demands, and data security, the study proposed practical strategies to ensure equitable access and safe interactions within the Metaverse. Empirical findings from our pilot experiment demonstrated the framework’s effectiveness in improving engagement and skill acquisition, with broader implications for educational policy and competency-based, experiential learning approaches. Looking ahead, we advocate for ongoing research to validate long-term learning outcomes and technological advancements to make immersive learning more accessible and secure. Our perspective underscores the transformative potential of the Metaverse classroom in shaping inclusive, future-ready educational environments capable of meeting the diverse needs of learners worldwide.

This paper envisions an innovative framework to tackle challenges in modern education through a comprehensive Metaverse classroom model. Drawing on immersive learning theories and filling critical gaps in current virtual learning approaches, the study structured a Metaverse-driven Multi-layered Immersive Learning Environment (Meta-MILE). The Meta-MILE framework integrates essential elements such as Metaverse infrastructure, personalized interactions, social collaboration, and advanced assessment techniques. The Meta-MILE model emphasized immersive Three-Dimensional (3D) environments, Artificial Intelligence (AI)-driven personalization, gamified learning pathways, and scenario-based evaluations to boost engagement and provide customized learning experiences. In light of potential challenges, we explored issues around accessibility, infrastructure demands, and data security, suggesting practical strategies to ensure equitable access and secure interactions. Empirical findings underscored Meta-MILE’s effectiveness in enhancing engagement and skill acquisition, and the discussion addressed educational policy implications and the role of immersive classrooms in moving toward competency-based, experiential learning. Future research recommendations included longitudinal studies to validate learning outcomes and continued technological innovation to make virtual learning more accessible and secure. Ultimately, this study highlighted the transformative potential of the Metaverse classroom in creating inclusive, future-ready educational environments tailored to 21st-century needs.

Integrating digital twins (DTs) into smart grid systems within the Internet of Smart Grid Things (IoSGT) ecosystem presents both security challenges and novel opportunities. Advanced Deepfake technologies enable adversaries to create highly realistic yet fraudulent DTs, threatening critical infrastructures’ reliability, safety, and integrity. In this paper, we introduce Authenticating Networked Computerized Handling of Representations for Smart Grid security (ANCHOR-Grid), a novel authentication framework that leverages Electric Network Frequency (ENF) signals as real-world anchors to secure smart grid DTs against Deepfake attacks. By capturing distinctive ENF variations from physical grid components and embedding these environmental fingerprints into their digital counterparts, ANCHOR-Grid provides a robust mechanism to ensure the authenticity and trustworthiness of virtual representations. We conducted comprehensive simulations and experiments within a virtual smart grid environment to evaluate ANCHOR-Grid. We crafted both authentic and Deepfake DTs of grid components, with the latter attempting to mimic legitimate behavior but lacking correct ENF signatures. Our results show that ANCHOR-Grid effectively differentiates between authentic and fraudulent DTs, demonstrating its potential as a reliable security layer for smart grid systems operating in the IoSGT ecosystem. These findings highlight the importance of integrating real-world environmental data into authentication processes for critical infrastructure and lay the foundation for future research on leveraging physical-world cues to secure digital ecosystems.

The Internet of Smart Grid Things (IoSGT) architecture, which incorporates cutting-edge Internet of Things (IoT) and Machine Learning (ML) technologies, is emerging as an effective enhancement for smart grid applications. By embracing the traditional Edge-Fog-Cloud hierarchy, IoSGT markedly advances the capabilities of existing infrastructure. The strategic placement and scheduling of Energy Storage System (ESS) capacities are vital for improving the performance, resilience, and robustness of IoSGT environments. To mitigate peak loads, optimize power profiles, and reduce system losses, this paper introduces an ELOCESS (Effective Location and Optimal Capacity of ESS) framework within the IoSGT architecture. The ELOCESS framework, which employs Time Domain Power Flow (TDPF) and Complex-Valued Neural Networks (CVNN), provides precise voltage fluctuation forecasts. The Plant Growth Simulation Algorithm (PGSA) determines the optimal storage locations and capacities. Our thorough simulation on a standard 33-node network proves that the ELOCESS method works to find the best places and sizes for battery storage systems, with an impressive 97.32% success rate for optimization.