Transformasi Pola Kejahatan Siber Berbasis Kecerdasan Buatan dan Implikasinya terhadap Efektivitas Sistem Penegakan Hukum

Abstract

The rapid advancement of digital transformation has significantly reshaped the landscape of cybercrime, particularly through the integration of Artificial Intelligence (AI), which enhances the automation, scalability, and sophistication of cyber attacks. This study aims to analyze the transformation of AI-based cybercrime patterns and examine its implications for the effectiveness of law enforcement systems, especially in developing countries. A mixed-method approach with a sequential explanatory design was employed, combining quantitative analysis of global cybercrime datasets from 2018 to 2025 with qualitative insights from 25 experts, including law enforcement officers, cybersecurity analysts, and academics. The quantitative findings indicate that AI-driven phishing dominates cybercrime activities, reaching 35% in Asia and 32% in America, while deepfake fraud accounts for up to 30% in Europe. The average investigation time for AI-based cybercrime increased from 60 days in 2018 to 180 days in 2025, three times longer than traditional cases. Meanwhile, conviction rates declined significantly, from 65% to 50% in Europe and from 58% to 42% in Asia. The study also reveals a widening technological gap, with AI adoption among offenders rising to 80% compared to 45% among law enforcement agencies in 2025. These findings highlight that the growing complexity of AI-based cybercrime directly reduces the effectiveness of law enforcement due to limitations in digital forensic capacity, regulatory adaptation, and cross-border coordination. This study contributes by integrating technological transformation analysis with institutional performance evaluation, offering a comprehensive perspective on cybercrime and law enforcement dynamics in the AI era.