The escalating cyber threats in the digital era, particularly within government institutions, underscore the urgent need to enhance compliance with cybersecurity policies (CCP). This study investigates how Coping Appraisal (CA) and Threat Appraisal (TA) shape CCP and Protection Motivation Intention (PMI) among Bangladeshi government employees, with Technology Anxiety (TEA) as a moderating factor. Grounded in Protection Motivation Theory (PMT), Technology Threat Avoidance Theory (TTAT), and General Deterrence Theory (GDT), the research adopts a quantitative approach, surveying 500 government employees across various hierarchical levels. Using Partial Least Squares Structural Equation Modeling (PLS-SEM), the study finds that CA significantly influences PMI and CCP, while TA exerts a stronger effect on CCP. TEA plays a dual moderating role, amplifying TA's impact while dampening CA's influence. The findings highlight the interplay of psychological, organizational, and technological factors in driving CCP. Practical recommendations include tailored training programs, tools to reduce technology anxiety, and culturally adapted policies to foster sustainable cybersecurity compliance. This research provides actionable insights for building resilient public-sector cybersecurity frameworks in resource-constrained, hierarchical contexts.
The research paper focuses on the implications of the development of the Dark Web on sexual offenses, with special emphasis on child sexual exploitation and trafficking. This qualitative study analyzes the Indian legislative framework, which includes the Information Technology (IT) Act, the Protection of Children from Sexual Offenses (POCSO), and the Bharatiya Nyaya Sanhita (BNS). The challenges identified are the lack of legislative measures to address the Dark Web, challenges in law enforcement and forensics, and the jurisdictional challenges. The recommendations include the development of cross-border legislation and technical measures in digital forensics. It also focuses on the role of digital forensic tools and the challenges to civil liberties, along with the importance of public-private partnerships. The study points to the gaps in the Indian legislative framework in comparison with the international standards, with emphasis on the need to achieve better victim-centered outcomes in investigations and the need to address the unique issues associated with the Dark Web.
Cybercrime and artificial intelligence are increasingly reshaping both the nature of crime and the methods used to study it. Understanding how these developments emerge within criminological scholarship is important for anticipating future directions in applied security research. This study examines thematic and methodological trends in 16,659 paper abstracts submitted to the annual meetings of the American Society of Criminology between 2014 and 2024. Conference abstracts offer a useful vantage point for detecting early-stage research agendas because they capture emerging topics and methodological innovation before journal publication. Using computational text analysis and topic modeling, the study identifies patterns of continuity and change in criminological research during a decade marked by rapid digitalization and the disruptions associated with the COVID-19 pandemic. The findings reveal two parallel developments: traditional areas such as criminological theory, policing, and courts remain central, while cybercrime and artificial intelligence are emerging as increasingly prominent frontiers. In particular, cybercrime research accelerates after 2020 and increasingly emphasizes hacking, malware, and data breaches. These findings suggest that criminology is undergoing a significant transition in which digital threats and computational methods are becoming more central to the study of crime and justice.
Crime rates often change abruptly across the borders of adjacent communities, yet the mechanisms producing these localized disparities remain poorly understood. This study develops a boundary centered analytical framework to examine how roadway network permeability and boundary land use context jointly shape crime rate differences between neighboring areas. Using incident level crime data from six US cities, we apply crime informed community detection to identify approximately 2,000 adjacency boundaries and analyze crime disparities using nonlinear descriptive models, regressions, and SHAP assisted machine learning diagnostics. Results reveal a nonlinear relationship between permeability and crime disparities in which gaps are largest under low connectivity, narrow rapidly at moderate connectivity, and weaken or diminish at high connectivity. Infrastructure barriers and vacant or industrial land create boundary vacuums associated with persistent and volatile violent crime disparities, whereas commercial corridors amplify property crime disparities through increased opportunity exposure. Overall, violent crime patterns reflect insulation produced by physical and land use barriers, while property crime patterns reflect exposure to target rich environments. These findings position urban boundaries as a critical ecological unit for understanding micro scale crime divergence and offer actionable insights for place-based security interventions, resource allocation, and applied models linking to localized crime dynamics.
This paper examines how adopting remote work affects crimes. Using mobility data as a proxy for residential occupancy and monthly crime administrative data across districts, we employ a local projections methodology to estimate the dynamic effects of remote work on different types of property crimes. Our identification strategy leverages temporal and spatial variation in remote work adoption patterns in Argentina, controlling for epidemiological, economic, and climatic conditions, district-specific characteristics, and common temporal shocks. We found that a one percentage point increase in the incidence of remote work is associated with an increase of up to 1.4 additional crimes in each district and month, with effects concentrated in residential and street robbery. The impact is greater in high-income districts. These findings contribute to our understanding of how changes in mobility patterns affect criminal behavior and have important implications for urban security policies in the context of evolving work arrangements.
With rapid advancements in online technology, police departments face increasing challenges and resource disadvantages in controlling cybercrime. Paradoxically, these same technological advancements provide a critical means for bringing offenders to justice. This exploratory study examines real-world police-guardian interactions to understand how Routine Activities Theory's concept of capable guardianship can be effectively applied to combat crime in the digital realm. Through a rigorous content analysis of publicly documented cases involving online crimes against vulnerable victims, specifically children and animals, this study identifies five key factors crucial to successful police-guardian interventions. These factors are: prestige gained by the capable guardian within their community, the perceived defenselessness of the victim, the anonymity afforded to the guardian, the strategic use of media amplification, and the level of trust established between the guardians and law enforcement. The findings offer a vital framework for policy makers and law enforcement to better understand, engage, and leverage online citizen groups, maximizing their collaborative potential to enhance security and improve offender apprehension in cyberspace.
This exploratory research examined the level of support and correlates of public support for using Facial Recognition Technology (FRT) to prevent credit card fraud. Based on the results from a Pew Research Survey Trends on the American Trends Panel, the study investigated two questions. First, what is the level of public support for using FRT to verify the identity of credit card holders? Second, which factors are most influential in supporting FRT use in retail environments? Our preliminary study demonstrated modest support for FRT in preventing credit card fraud and identified multiple factors that either positively or negatively affected support for its use. Black respondents showed less support for FRT use in retail settings than other racial/ethnic groups. Older male respondents, and more active internet users, were more supportive of FRT use to prevent credit card fraud. Conversely, those with liberal political leanings and reporting higher levels of education were less supportive of FRT use. We conclude by advocating for more research on public support for FRT in private spaces.
This study aims to identify key issues affecting the effectiveness of legal responses to the unlawful circulation of personal data, in order to improve the corresponding enforcement tools. The analytical framework includes a comparative analysis of approaches in the United States, the European Union, and Kazakhstan, while also accounting for the global reason of the problem. The research findings indicate that, under the current conditions of rapid digitalization, traditional criminal law mechanisms for combating the unlawful circulation of personal data demonstrate limited effectiveness due to both regulatory and technological barriers. Among the primary challenges are: the lack of a unified definition of criminal acts involving the processing and dissemination of personal data; difficulties in proving criminal intent; and the blurred boundaries between administrative and criminal offenses. These issues are compounded by technological difficulties in detecting violations, as well as the cross-border nature of digital crimes, which constrain the jurisdiction of national criminal justice systems. Moreover, existing criminal law provisions are often not adapted to the new forms of data processing and circulation. The study proposes several recommendations for improving criminal law regulation in the Republic of Kazakhstan within the relevant legal domain.
This study examined whether the frequency, success, success rates, and trends of terrorist attacks across various target types (e.g., abortion-related, infrastructure/services, business, educational institutions, government [diplomatic], government [general], journalists and media, military, police, private citizens and property, religious figures and institutions) changed before and after the 9/11 attacks in the U.S. The data consisted of monthly terrorism incidents in the U.S. from July 1981 to 2020, which were obtained from the Global Terrorism Database (N = 462). The negative binomial regression analysis indicated significant increases in attacks against religious figures and institutions, private citizens and property, and police, and significant decreases for government-diplomatic, abortion-related, business, and military targets. Successful attacks followed similar patterns, except for a decline in government-general targets and no change for military targets, while success rates mirrored these trends except for police, which showed no pre-post difference. ITSA results further showed significant post-9/11 increasing trends in attacks and successful attacks against private citizens and property (though not success rates) and against religious figures and institutions (including success rates). The findings of the study and their policy implications are discussed.
Active shooter incidents (ASIs) represent a significant and escalating threat to public safety, demanding robust, evidence-informed mitigation strategies for security practitioners and policymakers. This state-of-the-art scoping review systematically analyzes 223 peer-reviewed studies (2000-2024) to synthesize applied strategies across ASI prevention, operational response, and security-focused environmental design. Key findings highlight the critical roles of proactive threat assessment, Crime Prevention Through Environmental Design principles for secure environments and effective evacuation, multi-modal training (including virtual reality simulations), and coordinated multi-agency response protocols. The review also examines the utility of modeling techniques for enhancing preparedness and response effectiveness, emphasizing practical insights over technical complexities. Identified gaps include the need for more research on civilian response efficacy and the integration of emerging technologies into security management. This review provides an evidence-based framework and actionable recommendations to inform the development and implementation of effective security measures, policies, and crime prevention initiatives aimed at mitigating the impact of ASIs in diverse settings.
Trauma-informed, child-friendly investigative interviewing is essential to achieving justice while protecting child victims from further harm. This practitioner-oriented article explores the implementation of the National Institute of Child Health and Human Development (NICHD) Protocol in Hungary, focusing on its integration into law enforcement education and practice. Drawing from an interview with M & aacute;rta Fekete, trainer, educator, and one leader of Hungary's first postgraduate certification program in forensic child interviewing, the article highlights the challenges and transformative potential of structured interview techniques. Fekete reflects on the systemic limitations of traditional police training, the fragmented nature of interdisciplinary teaching, and the resistance to adopting non-confrontational methods. Emphasizing the role of experiential learning and emotional self-awareness, the training prioritizes both technical proficiency and personal growth. The manuscript discusses the broader societal implications of trauma-informed interviewing, including reduced victim blaming, greater public trust, and improved long-term outcomes for children. By promoting a paradigm shift from adult-focused interrogation to structured, empathetic communication focusing on children, the NICHD-based training fosters a more just, child-centered criminal justice process. The article closes with a call to redefine investigative success: not solely by conviction rates, but by the dignity and psychological safety of child victims preserved throughout the legal process.
Local governments have increasingly turned to digital innovations to address unique local challenges. An increasing number of U.S. cities are utilizing social media notification systems to keep their citizens informed about crime prevention and community safety issues. However, there are significant variations in how these systems are utilized across different local communities. This research focuses on the following questions: Why do some cities leverage social media notification systems to tackle local challenges, while others fall behind? What are the key determinants of social media notification system utilization in local governments for crime prevention? Drawing on perspectives of civic engagement and community traits, this study examines whether factors such as civic engagement, crime trend changes, and the prioritization of digital technologies for public safety influence a city's use of social media notification systems for crime prevention. Using data from the 2023 Local Digital Innovation Survey with California city governments, hierarchical logistic regression analysis reveals that civic engagement and community traits significantly influence California city governments' utilization of social media platforms for crime notifications.
TikTok's widespread popularity in the United States, despite ongoing media reports highlighting user privacy risks and concerns, raises a critical question about the role of media awareness in shaping privacy concerns, cybersecurity risk perception, and privacy protection behaviors among American TikTok users. Using a privacy calculus perspective, this study surveyed 1,202 TikTok users recruited through Amazon Mechanical Turk and employed structural equation modeling to analyze relationships among the key variables. The findings reveal that media awareness significantly predicted privacy concerns and cybersecurity risk perceptions, indicating that users more aware of media coverage are more likely to recognize potential privacy risks. Moreover, privacy concerns, in turn, mediated the relationship between media awareness and both cybersecurity risk perceptions and privacy protection behaviors, suggesting an increased awareness of digital risks. Individual factors like privacy awareness, prior victimization, and gratification levels were significant predictors of privacy concerns. This study underscores the critical role of media in shaping public perceptions of digital privacy risks and highlights the need for balanced reporting to avoid sensationalism. Implications include promoting digital literacy, enhancing platform transparency, and addressing individual privacy needs to foster safer online behaviors and trust in social media platforms.
Mr. Byungtak Kang is the Chief Executive Officer of AI SPERA, a leading startup cybersecurity company specializing in AI-driven Cyber Threat Intelligence (CTI). With a distinguished career spanning both industry and academia, he has held leadership positions at Nexon Korea, Nexon America, and Neople, overseeing critical infrastructure and security operations. He also served as an adjunct professor at Korea University's Graduate School of Information Security, contributing to the academic development of the field. A Microsoft MVP in Developer Security and author of two books, he has been instrumental in advancing secure systems development. At AI SPERA, he leads the development of Criminal IP, an open-source intelligence platform designed for attack surface assessment and threat hunting. His interdisciplinary experience, merging hands-on engineering, executive leadership, and academic insight, offers valuable guidance and inspiration to emerging and future professionals at the intersection of AI and cybersecurity.
The abundance of data generated, processed and stored by wearable devices presents a promising opportunity for digital investigations, including civil, criminal and medical cases. This study examined smartwatch data as digital evidence, using Garmin, Samsung, and BoAt devices paired with a Redmi phone. Forensic tools, Cellebrite UFED4PC, Magnet Axiom, and Oxygen Forensic Detective, were employed for data acquisition and analysis. The digital forensic analysis identified various artifacts, including physical activity data, health metrics, location data, user information and device information, all of which can serve as digital evidence in investigations. Cellebrite achieved the most comprehensive data extraction, revealing physical activity, health, location and user information. All tools accessed Garmin app data, but only Cellebrite decrypted Samsung's activity data. Location data was found in the Garmin app, but was encrypted in Samsung Health app data. BoAt app data was partially complete, yet provided user and activity details. Overall, the study highlights the importance of forensic analysis in extracting electronic evidence from smartwatch companion apps.
Privacy-Preserving Machine Learning is a method for preventing data leakage in machine learning algorithms. This method provides different techniques to train ML model collaboratively without revealing private information. It involves protecting against malicious attacks aimed at obtaining confidential information and causing breaches in data. Unlike regular ML, FL trains locally and does not send data to a server. This review paper examines the privacy risk associated with both traditional ML and FL. We categorize the attacks based on data, model and communication vectors, analyze the effects in privacy sensitive domains such as healthcare. Additionally, it also evaluates advanced defence strategies including Homomorphic Encryption, Differential Privacy, Secure Multi-Party Computation, and anonymization, discussing their effectiveness, scalability and trade offs between privacy and model utility. Future research directions include integrating adaptive privacy methods, using explainable AI (XAI) to improve model transparency, and creating strong privacy frameworks for edge computing and IoT applications. Finally, this paper gives an in-depth look at how privacy problems might be addressed in ML and FL, assuring the ethical use of these technologies in real-world applications.
This research demonstrates that integrating artificial intelligence into cyber range platforms significantly enhances cybersecurity readiness for cyber-physical systems by improving threat detection accuracy, accelerating incident response, and enabling adaptive learning. We developed a hybrid AI-powered cyber range architecture combining cloud-based simulation for scalability with emulation-based components for physical system fidelity. The framework leverages LSTM and GRU networks trained on 2.6 billion security events and implemented using Python 3.9 with Keras/TensorFlow, optimized via Adam optimizer (90.0% accuracy vs. 85.9% for ADAMAX). Results revealed three critical advancements: 91.3% classification accuracy in detecting coordinated attacks, identification of 76/107 security vulnerabilities (71% success rate) with 89.47% concept recognition, and a 34% reduction in detection-to-mitigation time compared to conventional cyber ranges. While demonstrating superior performance in controlled environments (90.9% accuracy in patch validation), challenges persist in AI explainability-only 58% of cybersecurity professionals could interpret model decisions, underscoring the need for interpretable machine learning frameworks in operational deployments.
Cybersecurity is a critical concern in contemporary digital environments, especially within the context of complex, interconnected systems. This study presents a systematic review of the complexities and inconsistencies surrounding the use of cyber-related terminology. A two-phased approach. The first part entailed using the PRISMA model to find relevant material, which was analyzed using ATLAS.ti software during the second phase. The analysis reveals ambiguity in cyber-related constructs, such as 'cybersecurity' versus 'cyber security', which impacts the clarity of research, policy development, and organizational practices, including education and training. Additionally, the study identified 'cybersecurity' as a primary security concern, interconnected with secondary and tertiary constructs. These relationships, visualized through ATLAS.ti Sankey diagrams, provide insight into how cyber-related constructs; all interrelated within the broader cyber ecosystem and within the dataset used for the study. This research is interesting and relevant because it clarifies the inconsistent use of cyber-related constructs and thus, each narrative constructed around cyberspace and its security. These taxonomic clarifications are also useful additions to curricula offering education and training in cyber-related subjects. Furthermore, organizations delivering security and/or intelligence services, within the context of cyber-related functional applications, can use such clarification to enhance their education, training materials, and functional environments.
The deployment of new technologies in law enforcement is on the rise, yet this integration has sparked significant public concern. Robotics has been no exception, sparking significant media attention to its adoption, implementation, and subsequent public responses. Through a series of multivariate analyses, the current study identifies the primary concerns and positive expectations voiced across various aspects of the reporting outlet, including time, location, and political viewpoints, as reported in the news media. The findings indicate that concerns regarding privacy, the use of lethal force, and overall skepticism toward the police use of these devices must be addressed. The paper also offers several recommendations on addressing public concerns when considering and implementing robotic technology in public safety. Agencies considering deploying such devices must take into account public concerns by establishing strict regulations surrounding the situational deployment of the robots and their data collection practices and consider deploying robots with safeguard mechanisms.
The rising phishing attack frequency in recent years has caused significant cybersecurity issues and presented a significant challenge for organizations and security professionals worldwide. Although several phishing detection methods have been proposed, they usually miss and cannot stop these changing attacks. Existing research contains constraints such as zero-day attacks, shortened URLs, and hidden harmful content. Often, these constraints lead to notable false positive and false negative rates, hence undermining the dependability of present techniques. This study offers a thorough examination of phishing detection research done over the last 10 years, from 2015 to 2024. The study examines phishing attack types, detection techniques, including heuristic, list-based, machine learning, and deep learning. Moreover, it emphasizes the dataset employed, the benefits, and the drawbacks of each. Building on the proposed taxonomy, this comprehensive article looks at present phishing detection running in phishing environments across mobile, website, and email platforms. The major objective is to identify study gaps and recommend paths for future work that can produce stronger, adaptive, and accurate phishing detection systems able to oppose developing threats.
