Key Takeaways
- • Samford University Computer Science student Daniel Abbott has identified over 2,200 cryptocurrency wallet addresses linked to investment scams using an agentic AI approach, earning "Platinum Badge" status within the Intelligence for Good Scambusters program.
- • On May 4, 2026 alone, Abbott's AI system identified 232 previously unknown crypto wallet addresses connected to scams, with those addresses subsequently flagged for over $5 million in on-chain fraudulent transactions within a single day in May.
- • Intelligence for Good shares discovered crypto addresses with law enforcement, cryptocurrency exchanges, and blockchain analysis firms to enable documentation and future transaction blocking, creating operational impact across the crypto compliance ecosystem.
- • The Scambusters project, led by Samford University Professor Scott Crews, recruits students to identify crypto investment scam websites and extract wallet addresses using advanced AI methodologies, embedding cybersecurity research into academic curriculum.
Summary
Daniel Abbott, a Computer Science student at Samford University, has deployed an agentic artificial intelligence system to identify cryptocurrency investment scam websites and extract associated wallet addresses at scale. Operating within the Intelligence for Good Scambusters program under the guidance of Professor Scott Crews, Abbott has become the project's top contributor, identifying more than 2,200 crypto addresses to date and holding the program's "Platinum Badge" designation for highest volume discovery.
The operational impact of Abbott's work extends beyond academic exercise. On May 4, 2026, Abbott's AI system identified 232 previously unknown cryptocurrency wallet addresses connected to active scams. Those wallet addresses alone subsequently tracked over $5 million in on-chain fraudulent transactions, demonstrating the direct correlation between address identification and real-world fraud activity. This efficiency—converting address discovery into quantifiable fraud metrics—underscores the effectiveness of the agentic AI methodology applied to scam detection.
The Scambusters initiative represents a structured approach to crowdsourced cybercrime research. Intelligence for Good recruits academic faculty to engage students in identifying fraudulent crypto investment schemes and extracting wallet addresses using AI-driven analysis. Upon discovery, the organization distributes these addresses to law enforcement agencies, cryptocurrency exchanges, and blockchain analysis companies. This distribution model creates an information-sharing ecosystem designed to enable documentation of fraudulent addresses and facilitate transaction blocking at the exchange and protocol level.
Abbott's contribution highlights the institutional value of embedding advanced AI research into cybersecurity education while generating operationally relevant intelligence for fraud prevention. The scale of his discoveries—over $5 million in identified fraudulent transactions from a single day's work—illustrates both the magnitude of crypto scam infrastructure and the potential for AI-driven detection methodologies to disrupt scam operations at meaningful scale.
The operational impact of Abbott's work extends beyond academic exercise. On May 4, 2026, Abbott's AI system identified 232 previously unknown cryptocurrency wallet addresses connected to active scams. Those wallet addresses alone subsequently tracked over $5 million in on-chain fraudulent transactions, demonstrating the direct correlation between address identification and real-world fraud activity. This efficiency—converting address discovery into quantifiable fraud metrics—underscores the effectiveness of the agentic AI methodology applied to scam detection.
The Scambusters initiative represents a structured approach to crowdsourced cybercrime research. Intelligence for Good recruits academic faculty to engage students in identifying fraudulent crypto investment schemes and extracting wallet addresses using AI-driven analysis. Upon discovery, the organization distributes these addresses to law enforcement agencies, cryptocurrency exchanges, and blockchain analysis companies. This distribution model creates an information-sharing ecosystem designed to enable documentation of fraudulent addresses and facilitate transaction blocking at the exchange and protocol level.
Abbott's contribution highlights the institutional value of embedding advanced AI research into cybersecurity education while generating operationally relevant intelligence for fraud prevention. The scale of his discoveries—over $5 million in identified fraudulent transactions from a single day's work—illustrates both the magnitude of crypto scam infrastructure and the potential for AI-driven detection methodologies to disrupt scam operations at meaningful scale.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.