Securing Data at Rest: Using ML-Driven Personally Identifiable Information(PII) Detection and Privacy-Preserving Techniques

Modern Day Enterprises collect a vast amount of Personal Identifiable Information (PII) from different sources, including interactions with customers, online behavior, survey responses, user feedback, Internet of Things (loT) devices, smartphone usage, and activity on social networks. This information, which comes in structured, semi-structured, and unstructured formats, has been stored in sources like Databases, File shares, Data Lakes, Cloud storages, and Data warehouses hosted in Cloud and On-prem data centers. This data is crucial for tailoring decisions, offering businesses a competitive advantage, and delivering personalized experiences to stakeholders. However, with strict data protection laws like the California Consumer Privacy Act (CCPA), the General Data Protection Regulation (GDPR) in Europe, and Brazil's General Data Protection Law (LGPD) taking center stage, there's a pressing need for advanced methods to identify and secure PII that is stored at rest in On-Prem, Public cloud and SAAS systems. This study investigates Machine learning (ML), and artificial intelligence (AI) to find PII stored across different data formats automatically. Furthermore, it examines how encryption, anonymization and differential privacy can be applied to enhance the security of this data. These strategies ensure compliance with current data privacy laws and promote ethical data handling practices. The introduction of a comprehensive system that combines ML-based PII detection with these privacy-strengthening solutions offers a scalable, effective, and immediate way to address potential threats and enhance data safety. Simulations have proven this system's capability to significantly improve data security and reduce the risk of exposing sensitive information while it's at rest, thereby safeguarding the integrity of sensitive information.