IIIT-H security group exposes flaws in digital payments

HYDERABAD: As artificial intelligence (AI) and digital payments weave into everyday life, researchers in Hyderabad are asking a critical question: How secure are the systems behind them? At the International Institute of Information Technology, Hyderabad (IIIT-H), the Security and Privacy Research Group is probing hidden vulnerabilities in AI models, blockchain networks, and mobile applications to identify risks before they can be exploited.

Led by Ankit Gangwal, the group studies how digital systems fail — and how those failures can be prevented. “Our lives are increasingly online. Savings move through digital wallets, and our phones unlock with our faces. If the software behind these systems is compromised, the impact can be huge,”Gangwal told Deccan Chronicle.One major focus is AI security. Researchers are investigating threats such as “model poisoning,” where attackers inject malicious data into training sets, causing AI systems to behave incorrectly under certain conditions.

They are also exploring protective measures like AI watermarking, embedding hidden ownership signatures in models so developers can prove authorship if their technology is copied. Blockchain and cryptocurrency networks form another area of study. By analysing unusual transaction patterns, the team aims to detect fraud and manipulation. Mobile security is equally important: researchers test how apps handle permissions and whether background processes could quietly leaksensitive data such as location, contacts, or financial details.By mapping out how attacks occur and where systems can break down, the group’s work strengthens the digital technologies that millions rely on every day.