Introduction
In 2017, there have been 1579 serious data breaches, which have resulted in over 180 million records being exposed (“Annual Number”). The world’s largest hacking attacks, including the Chinese breach of Google in 2009, Office of Personnel Management Data Breech in 2014, presidential elections tampering in 2016, the WannaCry and Petya virus outbreaks in 2017, and the Equifax data breach, have managed to affect the lives of millions of people worldwide.
This statistic represents the importance of cybersecurity in the 21st century (Patil, 1). Globalization had brought forth the unification of various information hubs into huge databases, which process and store data about mechanisms, processes, as well as day-to-day activities.
One of the most prominent issues in modern cybersecurity is the slowness of adaptation to evolving cyber threats. With the promotion and expansion of the Internet of Things (IoT) technology, the amount of labor required to conduct and perform updates is much greater than the amount of labor required to devise a new way around the protective grid (Jing et al., 2481). It is the reason why hackers remain one step ahead of cybersecurity software developers.
Should a security breach happen, a culprit could steal valuable files, take control of a technological appliance, or cause harm in a multitude of different manners. This is a great threat to users, businesses, governments, and militaries around the world. The purpose of this paper is to evaluate potential avenues for cybersecurity to change from a reactive to a proactive approach against cyber threats.
Literature Review
There are two approaches to cybersecurity. One is proactive, based on actively seeking out potential breaches and selectively securing important assets, while the other is reactive, which is known as a perimeter defense. It is established by traffic control, firewalls, passwords and login sequences, and strict user control. It is what most companies use to keep their systems and databases safe. Benaroch (315) explores the possibility of establishing a more flexible system based on mitigation and active decision-making, which can significantly lower the costs of cybersecurity for an enterprise.
The researcher proposes three types of active approaches to cybersecurity: mitigation interventions, prevention interventions, and learning interventions. Prevention interventions are conducted via standard means of protection such as antiviruses or firewalls, mitigation interventions include shutting off accesses to important data and repairing the damage one has been dealt with, and learning interventions include simulations of cyber-attacks to find ways of eliminating or reducing their chances of success. Technical and software necessities include base security measures, IDS, backup servers, and personnel training (Benaroch 332).
Jing et al. (2481) analyze the vulnerability of mass-internet implementation technologies such as the Internet of Things (IoT). According to the article, IoT does not have any uniform coding standards that would be used by the majority of appliances. It makes designing a protection system difficult, as it would require to encompass dozens of different systems into its architecture. Different types of programming mean numerous types of weaknesses present. Jing et al. (2483) also state that many IoT appliances have difficulty receiving timely updates on their coding and security.
Because of this weakness, defenses become out of date very quickly. Lastly, different IoT appliances may have conflicting codes that make programs interfere with one another and creates potential weaknesses for hackers to exploit. Jing et al. (2497) suggest several ways of protecting flows of heterogeneous data for IoT, which are implementing unified coding patterns for different technologies, using WSN and RFID technologies to enhance the perception level of the security architecture, and intelligent transportation systems to protect the input and output of information.
Patil (2) suggests another possible solution to the proactive security strategy. Artificial intelligence can be used as a medium to facilitate internet security. With the advancements in AI, it is possible to implement it alongside WSN and RFID. AI can be used to run learning simulations to try and infiltrate the defense, try cause training crashes and breakdowns, and test the security systems for any flaws. Also, artificial intelligence would be capable of responding to threats proactively without having any input from human operators. It would enable the cybersecurity system to become more fluid and responsive. However, as Patil (5) states, the development of AI is a complicated and expensive process, making it viable only for large corporations, government databases, and the military.
Possible Solutions
Both sources dedicated to the security issues of IoT and various enterprises indicate that the uniformity of the coding language throughout the system helps increase its potential responsiveness against cyber threats, as the specialists would be able to update the systems much quicker and mitigate any long-standing damage left by the attack. However, implementing such would mean that the hackers would be able to infiltrate the system much deeper, should the programming architecture experience a critical flaw.
Implementing a unified coding system for IoT would be impossible without all the producers agreeing on a single programming language and architecture to follow, instead of developing their own. In order to succeed in introducing it, international standards for internet security would be necessary. As it stands, the very concept of IoT is relatively new, which contributes to the heterogeneity of it. However, as the international community realizes the threats posed by these attacks, cooperation would be possible.
Another suggestion involves implementing proactive, multilayered defenses in order to react to various security threats as they appear. According to Benaroch (232), these systems are as efficient, if not more cost-efficient than the defensive perimeter systems, as they activate only once an actual threat to the integrity of the database appears. However, this type of security system requires staff training, as some threats will not be sorted automatically. They are also more complex compared to the basic security systems.
It is recommended to use AI to complement the proactive defenses and use it to run screenings and checks of the existing systems. However, its use will be limited due to low levels of development and high costs of implementation. However, government agencies and the military could create such programs and utilize them to protect critical facilities.
Conclusions and Recommendations
The state of the world’s cybersecurity right now is relatively poor, which is evidenced by the number of successful breaches happening in 2011-2018. Companies implement new technologies and provide internet access to everything without establishing proper security measures first. The recommendations to improve the situation to protect complex heterogeneous flows of data, such as the IoT, would include proactive defenses, using homogeneous codes and languages, and implementing AI in cybersecurity. This requires a united approach and a consensus between nations to reduce the possibility of a hacker threat.
Works Cited
“Annual Number of Data Breaches and Exposed Records in the United States from 2005 to 2018 (In Millions).” Statista. Web.
Benaroch, Michel. “Real Options Models for Proactive Uncertainty-Reducing Mitigations and Applications in Cybersecurity Investment Decision Making.” Information Systems Research, vol. 29, no. 2, 2018, pp. 315-340.
Jing, Qi, et al. “Security of the Internet of Things: Perspectives and Challenges.” Wireless Networks, vol. 20, no. 8, 2014, pp. 2481–2501.
Patil, Pranav. “Artificial Intelligence in Cybersecurity.” International Journal of Research in Computer Application and Robotics, vol. 4, no. 5, 2016, pp. 1-5.