Authors :
Kennelyn S. Araneta; Nurfraida A. Julasbi; Syeddin Nadvi A. Masbud; Fathar A. Mohammad; Juljamar A. Mohammad; Giner A. Nur; Haniza S. Sajiron; Ericka A. Salahuddin; Omar S. Tiamwatt; Shernahar K. Tahil
Volume/Issue :
Volume 9 - 2024, Issue 12 - December
Google Scholar :
https://tinyurl.com/fnxcszpp
Scribd :
https://tinyurl.com/2cema6d9
DOI :
https://doi.org/10.5281/zenodo.14565177
Abstract :
The research study utilizes a structural basis
characterized by a multi-method approach, including the
use of documents and interviewing the experts and
practitioners of cybersecurity. This is a dual technique
that allows for the dissection of AI applications that are
being used to catch phishing violation. The study also
reveals that the primary AI methods are machine
learning, which is the process of applying mathematical
and statistical algorithms to the pattern of email data to
classify the incoming mail as legitimate or malicious. This
is accomplished through NLP, which is the capability to
look at the words and phrases to determine if there are
any suspicious actions, in email traffic, to detect if there
were any changes. Recharge shows that artificial
intelligence plays a crucial role in boosting the accuracy
and efficiency of phishing detection system to a great
extent. In the case of machine learning models, for
example, the bot can be trained on a huge dataset to help
identify low-levels signs of phishing attacks, which will
eventually reduce the time needed of recognizing and
reacting. Furthermore, NLP algorithms allow the
practitioners a more profound examination of language
used in phishing emails; thereby, systems are able to
identify not only the common templates but also the novel
attacks that may not have patterns. On the other hand,
the research also sees the obstacle faced in the application
of AI in phishing detection. A key worry is the flexibility
of cybercriminals, who are perpetually coming up with
new methods to get around automated security systems.
The hide-and-seek nature includes the requirement of
perpetual training and the updating of AI models,
ensuring that the models are indeed resistant to attacks
from new forms of threats. Moreover, the study finds that,
although AI can do the majority of the work in phishing
detection, it is not infallible. There may be these false
positives and a few false negatives on some of them, which
could clearly lead to disruptions in legitimate
communications or help the criminals get on board
unchecked. Ethical considerations also emerge as a
significant theme in the research. The reliance on AI for
cybersecurity raises questions about privacy, data
security, and the potential for bias in algorithmic
decision-making. The study emphasizes the importance of
transparency in AI systems and the need for
organizations to maintain a balance between automated
solutions and human oversight.
Keywords :
Phishing Attack, Artificial Intelligence and Cybersecurity
References :
- Eze, C.S., & Shamir, L. (2024). “Analysis and prevention of AI-Based Phishing Email Attacks.” Electronics. 13(10), 1839.
- Jahankhani, M., Mavridis, P., & Alazab, M. (2020). Phishing Attacks: A Survey of the current Trends and Future Directions Journal of Cybersecurity and Privacy, 1 (3), 152-171.
- Bertino, E., & Islam, N. (2018). Botnets and Internet of Things Security. Computer Security, 76, 77-89.
- Sharma, A., & Gupta, R. (2021). Machine Learning Techniques for Phishing Detection: A Review. International Journal of Information Security, 20(3), 329-346.
- Zhang, Y., & Wang, T. (2020). Deep Learning for Phishing Detection: A Survey. Computers & Security, 92, 101737.
- Sarker, I.H., & Sultana, N. (2021). A Survey on Phishing Detection Techniques: Current Status and Future Directions. Journal of Network and Computer Applications, 177, 102924.
- Wang, Y., & Li, J. (2020). Challenges and Opportunities in Phishing Detection Using Machine Learning. IEEE Access, 8, 124123-124134.
- Alzahrani, A., & Alhassan, I. (2023). The Role of Artificial Intelligence in Detecting and Preventing Cyber and Phishing Attacks. ResearchGate.
- Cavoukian, A., (2019). Privacy by Design: The 7 Foundational Principles. Information and Privacy Commissioner of Ontario.
The research study utilizes a structural basis
characterized by a multi-method approach, including the
use of documents and interviewing the experts and
practitioners of cybersecurity. This is a dual technique
that allows for the dissection of AI applications that are
being used to catch phishing violation. The study also
reveals that the primary AI methods are machine
learning, which is the process of applying mathematical
and statistical algorithms to the pattern of email data to
classify the incoming mail as legitimate or malicious. This
is accomplished through NLP, which is the capability to
look at the words and phrases to determine if there are
any suspicious actions, in email traffic, to detect if there
were any changes. Recharge shows that artificial
intelligence plays a crucial role in boosting the accuracy
and efficiency of phishing detection system to a great
extent. In the case of machine learning models, for
example, the bot can be trained on a huge dataset to help
identify low-levels signs of phishing attacks, which will
eventually reduce the time needed of recognizing and
reacting. Furthermore, NLP algorithms allow the
practitioners a more profound examination of language
used in phishing emails; thereby, systems are able to
identify not only the common templates but also the novel
attacks that may not have patterns. On the other hand,
the research also sees the obstacle faced in the application
of AI in phishing detection. A key worry is the flexibility
of cybercriminals, who are perpetually coming up with
new methods to get around automated security systems.
The hide-and-seek nature includes the requirement of
perpetual training and the updating of AI models,
ensuring that the models are indeed resistant to attacks
from new forms of threats. Moreover, the study finds that,
although AI can do the majority of the work in phishing
detection, it is not infallible. There may be these false
positives and a few false negatives on some of them, which
could clearly lead to disruptions in legitimate
communications or help the criminals get on board
unchecked. Ethical considerations also emerge as a
significant theme in the research. The reliance on AI for
cybersecurity raises questions about privacy, data
security, and the potential for bias in algorithmic
decision-making. The study emphasizes the importance of
transparency in AI systems and the need for
organizations to maintain a balance between automated
solutions and human oversight.
Keywords :
Phishing Attack, Artificial Intelligence and Cybersecurity