Keynote Speaker

 Naveed Ahmad，Prince Sultan University, Riyadh, Saudi Arabia  

He is arrayed with both Research & Development throughout his career. He is young and energetic, seeking to join an organization where he can contribute to uplifting the economic and social status of people through the intervention of emerging technologies. One aspect of his career is contributing to research by publishing journals, book chapters, and conference papers, as well as supervising PhD and MS students. The second aspect involves bringing in and managing R&D projects at the university to promote the culture of applied research. The third aspect is establishing international linkages with Turkey, China, and the UK for the mutual benefit of the countries in general and the university in particular. The fourth unique aspect is his involvement in consultancy projects with local and international funding agencies, corporate entities, and the government.

Title: AI Based Defense Against OTP Flooding Attacks in Digital Authentication Systems
Abstract: Attackers have changed the way they attack to take advantage of One Time Passwords (OTPs), which are used more and more in digital authentication systems for two-factor authentication and identity verification. One such rising threat is OTP flooding, also known as OTP bombing. In this assault, attackers automate sending repeated OTP requests to overload systems, bother customers, or change how downstream services work (such SMS gateways and billing systems).
When faced with advanced botnets, proxy networks, or low-and-slow assault patterns, traditional countermeasures like CAPTCHA, rate-limiting, and static rules are not enough. These risks not only stop services from working, but they also cost businesses a lot of money and hurt their reputations, especially in the fields of finance, healthcare, and e-commerce.
This session will show an AI-based hybrid strategy that uses machine learning and behavioral analytics to find and stop OTP flooding threats in real time. The answer goes beyond static rate limits by using anomaly detection, supervised learning, and temporal user profiling to create a real-time risk rating engine.