Article Review #1

Understanding the Use of Artificial Intelligence in Cybercrime

Introduction : 

The article Understanding the Use of Artificial Intelligence in Cybercrime, by Sinyong Choi, Thomas Dearden, and Katalin Parti, explores the intersection of artificial intelligence (AI) and cybercrime. Published in the International Journal of Cybersecurity Intelligence and Cybercrime, this article addresses the rising threat of AI-driven cybercrimes, such as deepfakes and social engineering attacks. The article emphasizes the growing need for academic and professional communities to examine how AI is exploited by cybercriminals.

Relation to Social Sciences
The study connects directly to criminology and sociology by exploring how A.I contributes to criminal behavior. It uses routine activity theory (RAT) and cyber-routine activities theory (Cyber-RAT) to understand how cybercriminals exploit AI for malicious purposes. The study links technological advances with broader social issues, like cybersecurity vulnerabilities, the ethical implications of AI, and the changing nature of cybercrime. Additionally, it highlights the need to consider AI’s impact on marginalized communities, as these groups may be disproportionately targeted by AI driven cybercrimes.

Research Questions or Hypotheses
The study raises important questions about how AI is misused by cybercriminals, particularly through creating deepfakes and social engineering attacks. The main hypothesis is that AI enhances the effectiveness of cybercrimes. The authors aim to assess the extent of this issue and its impact, stressing the need for preventive measures and greater awareness of how AI can be used to deceive and manipulate individuals.

Research Methods
The authors use a mixed-method approach, combining qualitative expert interviews with quantitative data on AI-driven malware. This methodology provides a comprehensive understanding of the relationship between AI and cybercrime, considering both theoretical frameworks and real-world data. This approach ensures that both the technical and human aspects of AI-enabled crime are explored in depth.

Types of Data and Analysis
The study uses qualitative data from expert interviews and quantitative data on AI-driven cybercrime. The authors analyze this data using thematic analysis and statistical techniques, drawing upon RAT and Cyber-RAT frameworks to examine the dynamics of cybercrime. These methods allow for a nuanced understanding of the complex nature of AI-related criminal activities.

Conclusion
This article highlights the increasing threat of AI-driven cybercrime and calls for a proactive approach to cybersecurity. It provides essential insights into how AI can be exploited for criminal purposes, urging further research and policy development to counter these emerging threats. As AI continues to evolve, it is crucial that preventative measures are put in place to mitigate the risks posed by AI-enabled cybercrime.

References
Choi, S., Dearden, T., & Parti, K. (2024). Understanding the use of artificial intelligence in cybercrime. International Journal of Cybersecurity Intelligence and Cybercrime, 7(2), 1-3. https://vc.bridgew.edu/ijcic/

Article Review #2

“Testing Human Ability to Detect ‘Deepfake’ Images of Human Faces”

Introduction
  This article review examines the study titled “Testing Human Ability to Detect ‘Deepfake’ Images of Human Faces” (Bray et al., 2023) published in the Journal of Cybersecurity. The study investigates the effectiveness of human perception in distinguishing real human faces from AI-generated deepfake images. This review explores the research questions, methods, data analysis, connections to social sciences, challenges for marginalized groups, and broader societal contributions.

Research Questions and Hypotheses
The study addresses the primary research question: How accurately can humans differentiate real   human faces from deepfake images? (Bray et al., 2023) The hypothesis suggests that human detection abilities are significantly limited, with deepfake images often being mistaken for real ones due to their high level of detail and realism.

Research Methods
The researchers conducted an experimental study in which participants were shown a series of real and AI-generated images. (Bray et al., 2023). They were asked to classify each image as either real or fake. The study also examined the influence of participant demographics, familiarity with deepfakes, and confidence levels in detection accuracy.

Data Collection and Analysis
Data was collected from a diverse participant pool, and statistical methods were applied to measure accuracy rates, response times, and confidence levels. (Bray et al., 2023). The study used machine learning-generated deepfake images from various sources to test recognition capabilities. Analytical techniques included comparative analysis between different demographic groups and statistical significance testing of detection accuracy.

Relation to Social Sciences
This research intersects with social sciences by examining cognitive biases, trust in digital media, and the implications of misinformation (Bray et al., 2023). The study highlights how psychological factors influence the ability to detect synthetic media and how deepfakes can shape public perception and decision-making in political, social, and economic contexts.

Challenges for Marginalized Groups
The study discusses how marginalized groups are particularly vulnerable to deepfake-related misinformation and identity fraud. (Bray et al., 2023). Additionally, the lack of access to deepfake detection tools can disproportionately affect these communities, making them more susceptible to manipulation and exploitation.

Contributions to Society
The study provides valuable insights into human vulnerabilities to deepfake technology, emphasizing the need for improved detection tools and public awareness. (Bray et al., 2023) It calls for better education on media literacy and the development of AI driven solutions to counteract misinformation. The research also suggests policy recommendations to regulate deepfake usage in sensitive areas such as politics and journalism.

Conclusion
The study highlights the challenges in detecting deepfake images and the increasing threat they pose in misinformation campaigns. (Bray et al., 2023). While AI-driven detection tools continue to advance, human awareness and education remain crucial in combating deepfake manipulation. A combined effort from technologists, policymakers, and educators is necessary to create effective countermeasures. Future research could explore how training programs enhance human detection abilities and reduce susceptibility to deepfakes, ultimately fostering a more resilient and informed society.References
Bray, S. D., Johnson, S. D., & Kleinberg, B. (2023). Testing Human Ability to Detect ‘Deepfake’ Images of Human Faces. Journal of Cybersecurity, 9(1), Article ID tyad011. Testing human ability to detect ‘deepfake’ images of human faces | Journal of Cybersecurity | Oxford Academic