Abstract
This research study examines the impact of Artificial Intelligence (AI) data poisoning on data privacy violations in AI-enabled banking chatbots, employing a qualitative approach grounded in AI, data privacy, and cybersecurity theories. Through qualitative grounded theory research approach, viewpoints were gathered from a group of IT professionals in the banking sector. The research uncovered the impact of AI data poisoning across different professional roles, ranging from direct breaches to indirect exposure. Key findings revealed a spectrum of mitigation strategies, from technical solutions to basic awareness and mixed responses regarding the impact on personally identifiable information (PII), underscoring the complexity of safeguarding customer data [1]. Despite potential limitations stemming from the rapidly evolving AI landscape, this study contributes valuable insights into effective strategies for mitigating AI data poisoning risks and enhancing the security of AI-enabled chatbots in banking. It highlights the critical importance of developing robust security measures to protect sensitive customer data against privacy violations.