Introduction
Recent research in the area of cybersecurity shows that North Carolina is more and more often affected by more sophisticated attacks that cause damage to numerous areas of human life, from business to education. Some of the recent cases include an attack on a school in the state covered by Gordon and the launch of a cybercrime hotline noted by Coble. Nevertheless, the involvement of the human factor in the discussion of cybercrimes also leaves room for an argument that artificial intelligence could be of help when predicting and preventing cybercrimes (Smith). This emerging topic has to be covered because it could be the key to explaining why North Carolina and other states could switch to the implementation of artificial intelligence instruments instead of capitalizing on human resources. The complexity of fraudulent attacks requires the local cybersecurity agencies to come up with more instruments that might showcase the strength of AI-based solutions and validate the idea described above.
Pro Arguments
The first argument intended to strengthen North Carolina’s approach to protecting people from cybersecurity threats is that artificial intelligence adds to the scalability of the systems where it is employed. The speed of functioning also increases drastically, which might help the machine-based agents follow certain transactions and highlight suspicious operations way before those might be discovered by their human counterparts (Alashi and Badi 104). The increasing pace of online processes has to be maintained by machines since they are never subject to bias or human errors. North Carolina’s administration could also implement a machine learning system in order to have its cybersecurity systems handle even more transactions at once. This kind of instant processing will never be achieved with the help of a call center or the work of a team of cybersecurity specialists who do not use groundbreaking tools. Computerized analytics can be rightfully considered to be the future of fraud prevention.
Accordingly, the second benefit that has to be pointed out when discussing the strengths of artificial intelligence is the possibility of reducing manual labor. Therefore, the increasing number of attacks during the pandemic could be handled with the help of machine learning and have the numerous human workers assigned to less crucial tasks that actually require human judgment (Jang-Jaccard and Nepal 984). In a sense, a single piece of software could substitute hundreds of employees without forcing the institution to lose any of its productive strength. High-level projects would be then completed by humans, and tons of low-level missions would be accomplished with the help of AI-based instruments, creating more room for research and development (Hill 10). The number of menial tasks is not going to diminish, so the state administration should make a rational decision regarding the utilization of artificial intelligence and machine learning and employ these to support existing tools are reduce the pressure put on human employees during the Covid-19 pandemic.
The unmatched benefit of instruments based on artificial intelligence is that all of them are free of biased analysis that is typical of conventional employees. The majority of novel approaches that feature human workers are going to be affected by numerous limitations in the future due to the inherent partiality that cannot be eradicated (Soni 3). Previous experiences work differently with AI-based instruments since there is no subconscious to affect the computerized analysis of online transactions and suspicious operations. The administration could experiment with artificial intelligence and its derivatives in order to see how human workers compare to their digital counterparts. Human judgment will be forever flawed, but computers make no assumptions (Thesmar et al. 747). This should be the key to an accurate analysis assigned to a machine and not a team worker, even if the given human is rather capable and beyond knowledgeable.
Counter-Arguments
The first counter-argument that has to be considered is the high cost of implementation that is going to be affecting groundbreaking solutions at all times. If the State of North Carolina expects to benefit in full from the adoption of artificial intelligence, it will have to hire a complete team of data scientists. The latter would be responsible for developing an in-house system and setting up a production cycle allowing constant updates and revisions (Smith). The increasing amounts of data required to help the machine predict fraud also hint at the additional investments related to storage solutions. The popularity of the cloud and its comparatively safe architecture make it safe to say that the whole network space should be translated into the cloud as well, allowing the administration to step away from legacy systems.
Eventually, the administration of North Carolina would also have to consider the importance of finding a team of individuals who would possess an exceptional level of technical expertise. It would be required to build a complex, data-driven machine learning instrument that will have no significant flaws or gaps that might cause it to malfunction (Ryman-Tubb et al. 144). Under the condition where the implementation procedure itself might cost plenty of money, the development of the model would become even more expensive rather quickly. So as not to get exposed to any of the issues mentioned above, the administration would have to find experts who are closely linked to the notion of machine learning (Hill 10). These individuals should know exactly how to build a flawless iteration of a machine that can predict and prevent fraudulent network activities.
The ultimate counter-argument intended to prevent North Carolina’s administration from deploying an AI-based system to prevent numerous cyberattacks is the inability to collect effective data sets quickly. In other words, machine learning models are currently set up in a way that averts them from functioning if they do not have enough information on a certain threat (Soni 4). It is going to take the team an indeterminate period to cope with the challenges related to data collection and processing, with most end-users remaining without protection throughout that time. There are no easy ways to escape this challenge because of the growing number of transactions that are finalized daily. According to Zhu et al., without relevant information on all the transactions, the team will have no chance to prevent serious cyberattacks even with the help of artificial intelligence (74). The stage of proper machine training cannot be ignored as well, because the lack of experience would increase the number of false positives
Conclusion
Artificial intelligence can be deemed as one of the most important pillars of modern fraud detection. Therefore, human involvement in cybersecurity has to be reduced by a notch in order to help machines spot and destroy such threats without the involvement of any biased reviews or human-factor errors. In the age of the pandemic, the numerous administrations across the State of North Carolina should be willing to escape the principle of capitalizing on classic instruments and focus on the process of innovation. The number of false positives could be significantly reduced with the help of artificial intelligence, which makes it one of the most important digital trends that have to be adopted to slow down the progression of online fraud. Machine-based algorithms are going to be much more reliable than their human counterparts, affecting the overall quality of proposed solutions and reducing North Carolina’s exposure to such critical threats.
Works Cited
Alashi, Shahad, and Dhuha Badi. “The Role of Governance in Achieving Sustainable Cybersecurity for Business Corporations”. Journal of Information Security and Cybercrimes Research, vol 3, no. 1, 2020, pp. 97-112.
Coble, Sarah. “North Carolina Launches Cybercrime Hotline”. Infosecurity Magazine, 2020, Web.
Gordon, Brian. “Rise of Ransomware Attacks on NC Schools Hinder Virtual Learning”. Asheville Citizen Times, 2020, Web.
Hill, Charlotte. “Biometrics Becoming Must-Have for Fraud Prevention.” Biometric Technology Today, vol. 2018, no. 1, 2018, pp. 9-11.
Jang-Jaccard, Julian, and Surya Nepal. “A Survey of Emerging Threats in Cybersecurity.” Journal of Computer and System Sciences, vol. 80, no. 5, 2014, pp. 973-993.
Ryman-Tubb, Nick F. et al. “How Artificial Intelligence and Machine Learning Research Impacts Payment Card Fraud Detection: A Survey and Industry Benchmark.” Engineering Applications of Artificial Intelligence, vol 76, 2018, pp. 130-157.
Smith, Aaron. “Americans and Cybersecurity”. Pew Research Center, 2017, Web.
Soni, Vishal Dineshkumar. “Role of Artificial Intelligence in Combating Cyber Threats in Banking.” International Engineering Journal for Research & Development, vol. 4, no. 1, 2019, pp. 1-7.
Thesmar, David, et al. “Combining the Power of Artificial Intelligence with the Richness of Healthcare Claims Data: Opportunities and Challenges.” PharmacoEconomics, vol. 37, no. 6, 2019, pp. 745-752.
Zhu, Xingquan, et al. Fraud Prevention in Online Digital Advertising. Springer International Publishing, 2017.