The rise in cybercrime in Indonesia has prompted law enforcement agencies to optimize their preventive communication strategies based on visual content. This study conducts a comparative analysis of the use of two generative artificial intelligence platforms—Gemini (Google DeepMind) and Grok (xAI)—in the production of visual educational content by the Criminal Investigation Unit of the Banyumas City Police. The methodology employed is a comparative experimental research approach using identical prompt instruments across two main scenarios: the prevention of motor vehicle theft and the prevention of online fraud. The evaluation was conducted based on three assessment dimensions: contextual relevance, production speed (response time), and content filtering mechanisms. The study’s findings indicate that Grok outperforms Gemini in terms of production speed, the depth of local identity representation, and visual quality tailored to social media audiences, while Gemini demonstrates superiority in the dimensions of formality and consistency of output for the context of official institutional communication. The implications of this research point toward a recommendation for a complementary approach in the synergistic use of both platforms in accordance with the specific communication needs of the Criminal Investigation Unit.
Copyrights © 2026