The rapid growth of generative artificial intelligence (Gen-AI) is transforming industries such as healthcare, finance, education, law and software engineering. While Gen-AI enhances productivity and decision-making, it also introduces significant security, privacy and governance challenges, particularly through the unintended exposure of sensitive data.
Securing Uncontrolled Gen-AI Upload-Induced Data Leakage examines the risks associated with uploading personal information, intellectual property, confidential communications and proprietary source code to AI systems. It explores key threats, including unsafe prompt engineering, shadow AI usage, prompt injection, model inversion, data poisoning and API misuse, which can result in data leakage, model memorization and regulatory violations.
Through real-world case studies, forensic analysis and log tracing, this book demonstrates how sensitive information propagates across AI environments, and how such risks can be identified and mitigated. It also presents governance and forensic frameworks aligned with regulations such as GDPR, HIPAA and CCPA, providing practical guidance for secure, transparent and responsible AI adoption.
