GambitHunter: Indonesian Engineers Deploy AI to Combat Online Gambling
Jakarta – Online gambling remains a significant challenge in Indonesia, with platforms continually evading blocks by shifting domains and adopting diverse transaction methods. In response, a team of three Indonesian engineers has developed GambitHunter, an artificial intelligence (AI)-powered system designed to identify online gambling sites and trace associated transaction channels.
The system was created by Steven Sukma Limanus, Ilham Firdausi Putra, and Reynaldo Wijaya Hendry, and earned second place at the OpenAI Codex Hackathon in Singapore. The team completed the project in a remarkably short timeframe – just seven hours during the hackathon.
AI-Powered Site Exploration and Classification
GambitHunter distinguishes itself from manual investigation methods through its automated approach. An AI agent automatically explores the internet, searching for sites potentially linked to online gambling activities. According to Steven Sukma Limanus, “In the exploration stage, we try to find online gambling sites from several sources. There is an AI agent that automatically does this.”
Once a list of potential sites is compiled, GambitHunter employs an AI model to classify whether a site is genuinely an online gambling platform. Reynaldo Wijaya Hendry explained that the system utilizes OpenAI’s GPT-5.2 model for this classification process, enabling rapid analysis of numerous sites. “We use AI for this. So this exploration agent will try to collect as many lists of potential judol sites as possible. After that we use the OpenAI GPT-5.2 model to classify whether the website is really an online gambling site,” he said.
Automated Access and Data Extraction
To gain deeper insights, GambitHunter can create dummy accounts to access restricted areas of identified sites, such as deposit pages and payment method sections. This allows the AI to understand transaction flows. The system is capable of extracting payment information, including bank account numbers and images of payment pages. Reynaldo Wijaya Hendry noted, “Currently we can extract the image and account number used to receive the deposit.”
The team also suggests the system has the technical capability to interact with online gambling site customer service via chat, though this functionality hasn’t been fully tested.
AI’s Role in Development and Potential Impact
The developers emphasized the significant role of AI in both the development of GambitHunter and the application development process itself. Ilham Firdausi Putra highlighted the effectiveness of OpenAI Codex, stating, “Starting from designing architecture, aesthetics, to program code, everything uses Codex. I myself was surprised how effective and capable this Codex is.”
The team hopes GambitHunter will assist law enforcement agencies in mapping and disrupting online gambling networks. Steven Sukma Limanus stated their focus is “to facilitate parties interested in maintaining the digital ecosystem by finding and identifying online gambling transaction channels.” Ilham Firdausi Putra added that the system is ideally operated by a government agency with law enforcement authority, and the team is open to sharing code and knowledge.
During a demonstration at the hackathon, GambitHunter successfully identified real online gambling sites and extracted data from them, showcasing the potential of AI in combating illegal online activities.