For verification of this paper, please visit on http://www.socialresearchfoundation.com/shinkhlala.php#8

The unmatched speed of technological advancement has been significantly impacted by the incorporation of artificial intelligence (AI). AI is pervasive across many industries and has drawn praise and condemnation in equal measure. As it becomes a common element in the creation and operation stages of modern technologies, its expanding application offers both benefits and problems in cyber security. This paper provides a brief overview of artificial intelligence (AI) applications in cyber security and evaluates the possibility of strengthening defense mechanisms to improve cyber security capabilities. Upon examining the most recent artificial intelligence cyber security software, we may conclude that practical implementations are currently available. First, they are used with neural networks in a variety of distinct cyber security areas and the periphery. It was clear, therefore, that the use of artificial intelligence techniques would be necessary to effectively resolve some cyber security challenges. For example, comprehensive data is essential for making strategic decisions, and logical decision support is one of the unmet cyber security issues. Furthermore, it investigates AI-based models that either improve or jeopardize security across a range of cyber networks and infrastructures. In addition to discussing the socioeconomic effects of AI's engagement in cyber security, the study critically evaluates AI's role in creating cyber security applications and suggests ways to use upcoming technology to prevent risks and vulnerabilities created by AI.

The technology, procedure, and practice known as cyber security is used to protector against assaults, damage, and illegal admittance to networks, devices, programs, and information. "Cyber security refers to the set of activities and measures, technical and non-technical, intended to protect the'real geography' of cyberspace but also devices, software, and the information they contain communicated, from all possible threats," according to the explanation given by Myriam Dunn Cavelty (2). One of the most critical concerns in cyberspace these days is cyber security (3, 4). Numerous interdisciplinary interactions exist between artificial intelligence (AI) and cyber security. One way that artificial intelligence (AI) technologies, like deep learning, can be used in cyber security is by creating intelligent models for threat intelligence sensing, malware classification, and intrusion detection. Conversely, AI models will face a range of cyberthreats that will interfere with their ability to learn, make decisions, and sample. Therefore, in order to secure federated learning, avoid adversarial machine learning, and ensure machine learning privacy, specialist cyber security defense and protection solutions are needed for AI models. This learning offers a thorough analysis of AI applications in cyber security, including their benefits, challenges, and potential drawbacks.

The Evolution of Artificial Intelligence: Alan Turing, in Computing Machinery and Intelligence, published in 1950, introduced a mathematical approach and theory that laid the groundwork for the concept of artificial intelligence. But in 1949, computers could only execute commands; they could not store them. For this reason, a fundamental redesign of computers was required. Furthermore, computing was quite costly. These were the two main things preventing Turing from pursuing his goal. Then, Logic Theorist, a curriculum created by Allen Newell, Cliff Shaw, and Herbert Simon, approximated human problem-solving abilities and advanced the concept of artificial intelligence. This program is regarded as one of the earliest AI systems as well.

Although our understanding of artificial intelligence has grown over time, the fundamental problem remains unresolved: computers were not developed sufficient to execute intelligent programs. But with time, computers' processing and storage capacities increased to the point where Moore's law was no longer applicable. According to Moore's law, a computer's cost drops by half each two years as the quantity of transistors on a microchip doubles. When artificial intelligence (AI) achieved many of its objectives by the 2000s, it became a field in which everyone was interested (5).

History of cyber security- A software known as Creeper was created as part of an ARPANET research project. It traveled across the network and drop the communication "I'm the creeper, catch me if you can" anywhere it went. The creator of email, Ray Tomlinson, created a program named Reaper that monitored and eliminated Creeper. The earliest example of an antivirus was Reaper. The antivirus industry was undergoing significant development by the 1990s, and the first antivirus software was released in 1992. As soon as the globe went online, antivirus software and virus attacks became increasingly eminent. Antivirus programs similar to McAfee quickly entered the market and are still frequently used today. As more antivirus software entered the market in the 2000s, we can now find antivirus software on practically every device (6).

Why Should India Use AI for Cyber security?

Dynamic Threat Landscape: Cyber attacks are a persistent threat to India and the Asia-Pacific region. AI is ideal for proactively recognizing and countering these dangers because of its capacity for adaptation and learning.

Diverse Linguistic Environment: With a wide range of languages, AI is able to analyze data in diverse formats (text, audio, and video) and detect threats in various geographical areas.

Digital Transformation: AI protects sensitive data and vital infrastructure in real-time as India embraces digitalization.

AI-Powered Future Security- AI goes beyond threat detection to provide a future-ready cyber security framework for India. Block chain technology combined with artificial intelligence (AI) can provide tamper-proof data storage, making it very difficult for hackers to get past safeguards.

Strengthening Cyber security in India: Safeguarding Internal Threats

As India continues to embrace the 'Digital India' concept, the need for robust cybersecurity becomes increasingly critical. Here’s why strong cybersecurity is essential for protecting against internal threats in this evolving digital landscape:

1. Increased Digital Footprint: As digital infrastructure expands, so does the volume of sensitive data. This includes personal information, financial records, and critical government data. A strong cybersecurity framework is essential to protect this data from unauthorized access and potential breaches.
2. Internal Threats: Internal threats, including malicious insiders or negligent employees, pose significant risks. Cybersecurity measures must account for these threats by implementing strict access controls, monitoring systems for unusual activity, and fostering a culture of security awareness.
3. Economic Impact: Cyberattacks and data breaches can have severe economic consequences, including financial losses, reputational damage, and legal liabilities. Investing in robust cybersecurity helps mitigate these risks and safeguard the economic interests of both private enterprises and public institutions.
4. Protection of Critical Infrastructure: As India modernizes its infrastructure, including energy, transportation, and healthcare systems, the risk of cyberattacks targeting these critical sectors increases. Strong cybersecurity is essential to ensure the resilience and reliability of these vital services.
5. Regulatory Compliance: With the introduction of various data protection laws and regulations, such as the Personal Data Protection Bill (PDPB), organizations must comply with stringent cybersecurity standards. A well-defined cybersecurity strategy helps meet these legal requirements and avoid potential penalties.
6. National Security: Cybersecurity is crucial for national security. Ensuring that government communications, defense systems, and intelligence operations are secure from cyber espionage or sabotage is vital for maintaining national security.
7.  Public Confidence: As more citizens engage with digital services, their trust in these platforms depends on their security. Effective cybersecurity measures help build and maintain public confidence in digital transactions, e-governance, and online services.
8. Innovation and Growth: Strong cybersecurity supports innovation by providing a secure environment for technological advancements. It encourages the development of new digital solutions and services by reducing the fear of cyber threats.

To address these challenges, India needs to invest in several key areas:

1. Skilled Workforce: Developing a skilled cybersecurity workforce through education and training programs is essential for managing and mitigating internal threats effectively.
2. Advanced Technologies: Leveraging advanced technologies such as artificial intelligence and machine learning can enhance threat detection and response capabilities.
3. Collaboration and Information Sharing: Encouraging collaboration between government agencies, private sector organizations, and international partners can improve overall cybersecurity posture and facilitate timely responses to emerging threats.
4. Public Awareness: Promoting cybersecurity awareness among citizens and organizations helps create a culture of vigilance and responsible digital behavior.
5. By focusing on these areas, India can build a robust cybersecurity framework that protects against internal threats while supporting the goals of the 'Digital India' initiative.

Enhancing Cybersecurity in India: Defending Against Cyber Threats from different countries

As India continues to develop its digital infrastructure, it faces significant cybersecurity threats from various state actors, including China, North Korea, and Iran. Here’s why strong cybersecurity is crucial for protecting against these external threats:

Targeted Cyberattacks:

China has been linked to various cyber espionage activities, targeting sensitive information related to defense, technology, and strategic industries. These attacks often aim to steal intellectual property, disrupt operations, or gather intelligence. Similarly, North Korea is known for its cybercriminal activities, including ransomware attacks and financial theft. Their cyber operations often target financial institutions and critical infrastructure to generate revenue or create disruption. Iran has been involved in cyberattacks aimed at disrupting critical infrastructure and conducting espionage. Their attacks often target energy sectors and other vital components of a nation’s infrastructure.

Protection of Critical Infrastructure:

1. Energy and Utilities: Cyberattacks on energy grids and utilities can cause widespread disruptions. Ensuring robust cybersecurity measures can help prevent such attacks and ensure the continuous operation of critical services.
2. Transportation Systems: Transportation networks are vital for economic stability and public safety. Protecting these systems from cyber threats is essential for maintaining operational integrity and security.

National Security and Sovereignty:

1. Defense Systems: Cyberattacks targeting defense systems can compromise national security. Strong cybersecurity is crucial for protecting defense communications, weapons systems, and military operations from external threats.
2. Government Communications: Securing government communication channels helps prevent espionage and ensures the confidentiality of sensitive diplomatic and strategic information.

Economic Stability:

1. Financial Sector: Cyberattacks on financial institutions can lead to significant financial losses and undermine public confidence in the banking system. Ensuring the security of financial transactions and data is vital for economic stability.
2. Intellectual Property: Protecting intellectual property from theft and industrial espionage is crucial for maintaining competitive advantage and encouraging innovation.

Strategic and Tactical Response:

1. Threat Intelligence: Implementing advanced threat intelligence capabilities helps detect and mitigate potential attacks from state actors. This includes monitoring for signs of intrusion, analyzing threat patterns, and responding to emerging threats in real time.
2. Incident Response: Developing a robust incident response plan ensures a swift and effective reaction to cyberattacks. This includes having trained personnel and resources in place to handle and mitigate the impact of an attack.

Diplomatic and Geopolitical Implications:

1. International Cooperation: Collaborating with international allies and cybersecurity organizations can enhance defense against state-sponsored cyber threats. Sharing information and best practices helps strengthen collective cybersecurity efforts.
2. Strategic Alliances: Forming strategic alliances and participating in global cybersecurity initiatives can bolster India’s defenses and provide additional resources and support in addressing cyber threats.

Steps for Strengthening Cybersecurity:

1. Investment in Technology and Infrastructure: Upgrading cybersecurity infrastructure with advanced technologies like AI and machine learning can enhance threat detection and response capabilities.
2. Building Cybersecurity Expertise: Developing a highly skilled cybersecurity workforce through education and training is essential for managing complex threats and ensuring effective defenses.
3. Enhancing Threat Detection and Response: Implementing advanced threat detection systems and establishing a well-coordinated incident response strategy can help address and mitigate cyber threats.
4. Promoting Cyber Hygiene: Educating individuals and organizations about best practices for cybersecurity can help prevent vulnerabilities and reduce the risk of successful attacks.
5. Legislative and Policy Measures: Strengthening cybersecurity policies and regulations can provide a framework for protecting digital infrastructure and ensuring compliance with security standards.
6. By addressing these areas, India can build a strong cybersecurity posture capable of defending against sophisticated cyber threats from state actors and safeguarding its digital infrastructure effectively.

AI TECHNIQUES IN CYBERSECURITY-

An overview of learning algorithms, which are key concepts in artificial intelligence, is given in this section. It also gives a quick rundown of several of the AI subfields—expert systems, machine learning, deep learning, and biologically inspired computation—that are frequently applied in the subject of cyber security.

Machines can be skilled using learning algorithms, which also assist humans, execute improved by allowing humans to learn from their mistakes. "A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E," states Mitchel's definition (7). Three popular learning algorithms are used to educate machines, and they are explained below:

1. Supervised learning: This kind uses a sizable labeled data collection during the training phase. A test data set must be used to verify the system following the training phase. Typically, these learning algorithms are employed as regression or classification mechanisms. Based on the input, the regression method produces outputs, or prediction values, which are one or more continuous-valued numbers. Algorithms for classification divide data into groups and produce distinct results, unlike regression procedures.
2. Unsupervised learning: Unsupervised learning is different from supervised learning in that it uses an unlabeled training data set. Typically, unsupervised learning approaches are used to cluster data, estimate density, and reduce dimensionality.
3. Reinforcement learning: When faced with incentives or punishments, this type of algorithm learns the best possible behavior. Reinforcement can be conceptualized as a combination of supervised and unsupervised learning. Reinforcement learning can be useful when there is a lack of available data (8).

AI technology encompasses many subfields, some of which are included here.

The expert system or ES: It is often referred to as a knowledge-based system. An inference engine, which is used to reason over pre-set data and find answers to problems, and a set of knowledge, which serves as the foundation for knowledge-based systems and combines acquired experiences, are the two main components of ES (9). Proficient systems are capable of handling two different kinds of troubles: rule-based reasoning and case-based reasoning, according to the analysis technique. Cyberspace decision-making support can be provided by ESs. Generally speaking, altered security system data are assessed before the security expert system decides whether or not a system or network behavior is malicious. Security professionals typically employ statistical techniques to quickly scan and examine a huge amount of altered data. Expert systems that monitor in real time in cyber settings can effectively assist these efforts. Security experts can choose the proper security steps in the event of harmful intrusions by using the pertinent information and warning message generated by security expert systems (10).

Machine learning (ML): “Machine learning is a set of methods that gives computers the ability to learn without being explicitly programmed,” states Arthur Samuel's definition (11). Through machine learning (ML), systems can learn from data, find and formalize the underlying principles, and improve through experience without needing to be explicitly coded. In order to find patterns in the data and base future decisions on the provided examples, the learning process starts with monitoring the data through examples. The program can use this information to reason the properties of cases that haven't been seen before (12). ML makes use of statistics to analyze vast amounts of data in order to extract information, identify patterns, and make conclusions. Different kinds of machine learning algorithms exist. They can also be broadly divided into three groups: reinforcement learning, unsupervised learning, and supervised learning. The most often utilized algorithms in the field of cyber security are ensemble learning, k-means clustering, decision trees, support vector machines, Bayesian algorithms, k-nearest neighbor, random forests, association rule algorithms, and principle component analysis (13).

Deep learning (DL) - Deep learning is a subset of machine learning and another subset of artificial intelligence. The success of deep learning networks in tasks like speech recognition, computer vision, and self-driving cars has drawn a lot of attention to it in recent years.  Layers of connected processing nodes, or neurons, make up deep learning networks. A statement or an image from the outside world is fed into the first layer, also known as the input layer. After processing the input, the following layer transfers it to the following layer, and so forth. It's common to refer to these intermediary layers as hidden layers. When an object in a picture is recognized or a sentence is translated between languages, for example, the output layer produces a prediction or classification at the end.  The term "deep" refers to the several levels present in these networks. A network's depth is crucial because it enables the network to recognize intricate patterns in the input. By varying the strength of connections among the neurons in each layer, deep learning networks are able to learn how to carry out intricate tasks. We refer to this procedure as "training." The data used to train the network determines the strength of the connections. The network's ability to carry out the task it was trained to complete will improve with the amount of data used.

Biologically inspired computation: It It alludes to a collection of complex algorithms. and techniques that leverage biological traits and behaviors to address a variety of challenging issues. The ways that traditional AI and bio-inspired approaches learn differ from one another.
Conventional AI generates intelligence, which machines may display. Programs that produce other programs, including intelligence, are what make this intelligence. But the foundation of bio-inspired computing is made up of a few basic laws and the simple creatures that strictly follow them. Under some circumstances, these organisms undergo slow evolution. The bio-inspired computations that are most frequently employed in the cyber security field include the following methods: The cyber security field most frequently employs genetic algorithms, evolution strategies, ant colony optimization, particle swarm optimization, and artificial immune systems among other bio-inspired computations (13).

Case Studies Highlighting AI's Contribution to Cyber security- The potential of artificial intelligence to support cyber security is being more widely acknowledged and utilized by enterprises globally. This is demonstrated by the numerous real-world applications that have employed AI-driven tools and solutions to strengthen defenses, improve threat recognition, and supervise vulnerabilities. Here, we examine a few notable case studies that show how artificial intelligence is transforming the cyber security industry.

A.     Symantec’s targeted attack analytics (TAA) tool- The Targeted Attack Analytics (TAA) product from Symantec One of the most notable examples of artificial intelligence (AI) in the cyber security space is Symantec's Targeted Attack Analytics (TAA) product. This cutting-edge application uses artificial intelligence (AI) to automatically analyze enormous volumes of data and spot signs of a security compromise. TAA makes use of sophisticated AI algorithms that replicate the procedures, analysis of data, and roles of seasoned security professionals. TAA can accurately identify targeted assaults by "learning" from human experts (14, 15).

When it came to fending off a Dragonfly 2.0 attack in 2018, TAA proved to be efficient at spotting and neutralizing advanced threats. This incident demonstrated the tool's capacity to handle incidents and proactively identify risks, greatly enhancing the effectiveness of cyber security responses. Proactive threat identification and incident management have advanced significantly with the integration of AI in solutions like TAA (16). Cyber security experts are able to significantly increase the overall security of their systems by using AI to better prevent and respond to specific attacks.

B.    Sophos’ intercept XSophos’ Intercept X tool- In the area of cyber security, it is a potent use of artificial intelligence (AI). With the use of deep learning neural networks, which are based on the functioning of the human brain, this sophisticated technology can accurately discriminate between benign and dangerous files. In a matter of milliseconds, Intercept X can evaluate thousands of features from a file, carry out a thorough analysis, and conclude if the file is safe or possibly dangerous (17-19).

The system has a high degree of accuracy in identifying both known malware and zero-day threats because it was trained using real-world feedback and two-way threat intelligence. Furthermore, Intercept X keeps its false-positive rate low, reducing the possibility of mistakenly classifying benign files as dangerous. This case study illustrates how artificial intelligence (AI) may strengthen defenses against cyber attacks. Tools like Intercept X can greatly increase system security by utilizing AI to add accuracy and agility to malware detection and threat prevention (20-22).

C.    IBM’s QRadar advisor with Watson- By exploiting AI in cyber security by means of its QRadar Advisor tool, IBM has made notable strides. This program automatically looks into possible security incidents using IBM Watson's cognitive computing capabilities. The QRadar Advisor can help security analysts evaluate threat situations and lower the chance of missing serious attacks by utilizing AI.In this instance, using AI to threat detection and response increases accuracy and efficiency. The QRadar Advisor can swiftly evaluate massive volumes of data and accurately identify any dangers thanks to sophisticated AI algorithms. By giving security analysts strong capabilities to identify and address cyber risks, this eventually improves an organization's cyber security architecture (23, 24).

D.    DeepLocker- Artificial intelligence (AI) has been used both maliciously and to enhance cyber security, as demonstrated by the development of DeepLocker, a new type of malware driven by AI. DeepLocker is much harder to identify and stop than typical malware because, unlike traditional malware, it may hide its dangerous intent until it targets a particular victim. This cutting-edge virus precisely identifies its victim by using artificial intelligence (AI) and indicators like face recognition and relocation. This instance demonstrates how AI in cybersecurity has two drawbacks. Artificial intelligence (AI) has the potential to strengthen cybersecurity defenses, but it may also be used to produce sophisticated malware that is challenging to identify and remove. This emphasizes how crucial it is to keep developing AI-powered cyber security tools and precautions to prevent AI from being used maliciously. (25, 26).
These case studies show how artificial intelligence (AI) may significantly improve cyber security when used properly. But it also highlights the need to stay up to date with the changing threat landscape, as seen by the DeepLocker incident, where AI itself might be weaponized. Artificial intelligence (AI) is becoming a key component in the development of strong and resilient cyber security systems, from automating the detection of sophisticated threats to learning from real-time data for proactive protection.

Challenges in Intelligent cyber security

Future study, development, and use of AI in cyber security will necessitate the ability to discern between short- and long-term goals. There are many AI applications that are directly relevant to cyber security, and there are current cyber security challenges that require more clever solutions than are already in place. We haven't yet discussed these current, practical uses. Promising developments are anticipated when it comes to the application of completely new data handling concepts in state of affairs management and decision-making. These guidelines support the implementation of a standardized, hierarchical data design in the software used for decision-making. This kind of layout has been organized. One challenging application area is internet central warfare data management. The only way to provide quick state-of-affairs assessments that give decision makers and leaders an advantage over others at any C2 level is through autonomous data management. Intelligent systems are currently being utilized in a number of applications; they are usually concealed within a program, such as in software security measures. On the other hand, if large databases are created, intelligent systems will find greater use. This might need a substantial investment in the creation of massive standard data bases and data collecting. Given the far-off future—at least a few decades from now—perhaps we shouldn't constantly forbid us from using "narrow AI." Some people firmly believe that by the middle of the current century, artificial general intelligence—the main aim of AI development—will have been accomplished.

In a world where malicious software and cyberattacks are becoming more frequent, advanced cyber security methods are crucial. Furthermore, DDoS avoidance experience has demonstrated that security against massive attacks may be achieved with relatively little resources if innovative approaches are used. Analyses of published works indicate that the most broadly relevant AI findings for cyber security come from research on artificial neural networks. Neural network applications for cyber security are continuously being developed.
In many fields where neural networks are not the best fit technologies, sophisticated cybersecurity methods are nevertheless necessary. These fields include decision support, scenario comprehension, and information control. The most interesting part of this scenario is expert machine development.

Although the rate at which universal artificial intelligence will advance is impossible to forecast, it is nevertheless possible that individuals who carry out these crimes may utilize any newly developed AI. This is not presumable. Furthermore, systems' cybersecurity capabilities would be significantly improved by the most current technical developments in information management, interpretation, and understanding—particularly in the area of computer learning.

1. Josh Fruhlinger, “What is cyber attack?,”. CSO, February 2020.
2. Cavelty, M. D., Mauer, V., & Balzacq, T. (Eds.). (2010). The Routledge handbook of security studies (No. s 56). London: Routledge.
3. Guan, Z., Li, J., Wu, L., Zhang, Y., Wu, J., & Du, X. (2017). Achieving efficient and secure data acquisition for cloud-supported internet of things in smart grid. IEEE Internet of Things Journal, 4(6), 1934-1944.
4. Wu, J., Dong, M., Ota, K., Li, J., & Guan, Z. (2018). Big data analysis-based secure cluster management for optimized control plane in software-defined networks. IEEE Transactions on Network and Service Management, 15(1), 27-38..
5. Haenlein, M., Kaplan, A., Tan, C. W., & Zhang, P. (2019). Artificial intelligence (AI) and management analytics. Journal of Management Analytics, 6(4), 341-343.
6. Warner, M. (2012). Cybersecurity: A pre-history. Intelligence and National Security, 27(5), 781-799.
7. Tom M. Mitchel, “Machine Learning,”. McGraw-Hill Science/Engineering/Math; March 1997, ISBN: 0070428077.
8. Arulkumaran, K., Deisenroth, M. P., Brundage, M., & Bharath, A. A. (2017). Deep reinforcement learning: A brief survey. IEEE Signal Processing Magazine, 34(6), 26-38.
9. Wirkuttis, N., & Klein, H. (2017). Artificial intelligence in cybersecurity. Cyber, Intelligence, and Security, 1(1), 103-119.
10. Benjamin, D. P., Pal, P., Webber, F., Rubel, P., & Atigetchi, M. (2008, August). Using a cognitive architecture to automate cyberdefense reasoning. In 2008 Bio-inspired, Learning and Intelligent Systems for Security (pp. 58-63). IEEE.
11. Samuel, A. L. (1959). Some studies in machine learning using the game of checkers. IBM Journal of research and development, 3(3), 210-229.
12. Chumachenko, K. (2017). Machine learning methods for malware detection and classification.
13. Truong, T. C., Diep, Q. B., & Zelinka, I. (2020). Artificial intelligence in the cyber domain: Offense and defense. Symmetry, 12(3), 410.
14. Liagkou, V., Stylios, C., Pappa, L., & Petunin, A. (2021). Challenges and opportunities in industry 4.0 for mechatronics, artificial intelligence and cybernetics. Electronics, 10(16), 2001.
15. Thiyagarajan, P. (2020). A review on cyber security mechanisms using machine and deep learning algorithms. Handbook of research on machine and deep learning applications for cyber security, 23-41.
16. Murdoch, B. (2021). Privacy and artificial intelligence: challenges for protecting health information in a new era. BMC Medical Ethics, 22, 1-5.
17. Ghazal, T. M., Hasan, M.K., Zitar, R.A., Al-Dmour, N.A., Al-Sit, W.T. and Islam S. (2022) “Cybers Security Analysis and Measure-ment Tools Using Machine Learning Approach,”2022 1st International Conference on AI in Cybersecurity, ICAIC 2022.
18. Kshetri, N. (2021). Economics of artificial intelligence in cybersecurity. IT Professional, 23(5), 73-77.
19. Abusamrah, I., Madhoun, A., & Iseed, S. (2021). Next-Generation Firewall, Deep Learning Endpoint Protection and Intelligent SIEM Integration.
20. Hamid, K., Iqbal, M.W., Aqeel, M., Liu, X. and Arif, M. (2023).“Analysis of Techniques for Detection and Removal of Zero-DayAttacks (ZDA),” pp. 248–262.
21. Teodorescu, C. A. (2022). Perspectives and reviews in the development and evolution of the zero-day attacks. Informatica Economica, 26(2), 46-56.
22. Kapoor, A., Gupta, A., Gupta, R., Tanwar, S., Sharma, G., & Davidson, I. E. (2021). Ransomware detection, avoidance, and mitigation scheme: a review and future directions. Sustainability, 14(1), 8.
23. Das, R., & Sandhane, R. (2021, July). Artificial intelligence in cyber security. In Journal of Physics: Conference Series (Vol. 1964, No. 4, p. 042072). IOP Publishing.
24. Sasikala, D., & Sharma, K. V. (2022). Deployment of artificial intelligence with bootstrapped meta-learning in cyber security. Journal of Trends in Computer Science and Smart Technology, 4(3), 139-152.
25. Taddeo, M. (2019). Three ethical challenges of applications of artificial intelligence in cybersecurity. Minds and machines, 29, 187-191.
26. Yu, N., Tuttle, Z., Thurnau, C. J., & Mireku, E. (2020, April). AI-powered GUI attack and its defensive methods. In Proceedings of the 2020 ACM Southeast Conference (pp. 79-86).