Analisis Serangan Phising Dan Upaya Pencegahannya Pada Sistem Informasi

Phishing attacks continue to evolve into increasingly complex cyber threats, particularly with the emergence of Generative Artificial Intelligence (GenAI) technologies capable of producing realistic fraudulent content. This study highlights the ineffectiveness of traditional blacklist-based security methods in countering dynamic attacks that exploit both technical and psychological vulnerabilities of users. The objective of this research is to analyze the technical characteristics and masquerading patterns of current phishing attacks. Using a descriptive qualitative approach, this study examines 14 active phishing URL samples from the 2024-2025 period through URL Feature Analysis. The results indicate that 80% of the samples have adopted the HTTPS protocol to manipulate user trust. Furthermore, the study identifies a dominance of typosquatting techniques and the abuse of cheap Top-Level Domains (TLDs) that often bypass standard security detection. The research concludes that effective cyber defense requires the integration of adaptive strategies, combining intelligent Machine Learning-based detection systems with comprehensive security awareness training to mitigate risks associated with human error.