Search papers, labs, and topics across Lattice.
This paper introduces a secure-by-design GenAI framework, integrating PromptShield and the Cloud Investigation Automation Framework (CIAF), to enhance cloud security and forensics. PromptShield uses ontology-driven validation to defend LLMs against prompt injection attacks, while CIAF streamlines cloud forensic investigations through structured, ontology-based reasoning. Evaluations on AWS and Azure datasets demonstrate that PromptShield improves LLM classification performance under attack (93%+ precision/recall/F1), and CIAF enhances ransomware detection accuracy using Likert-transformed performance features.
Bolstering LLMs with ontology-driven input validation and structured forensic reasoning yields a secure-by-design GenAI framework that substantially improves both security and forensic accuracy in cloud environments.
As cloud environments become increasingly complex, cybersecurity and forensic investigations must evolve to meet emerging threats. Large Language Models (LLMs) have shown promise in automating log analysis and reasoning tasks, yet they remain vulnerable to prompt injection attacks and lack forensic rigor. To address these dual challenges, we propose a unified, secure-by-design GenAI framework that integrates PromptShield and the Cloud Investigation Automation Framework (CIAF). PromptShield proactively defends LLMs against adversarial prompts using ontology-driven validation that standardizes user inputs and mitigates manipulation. CIAF streamlines cloud forensic investigations through structured, ontology-based reasoning across all six phases of the forensic process. We evaluate our system on real-world datasets from AWS and Microsoft Azure, demonstrating substantial improvements in both LLM security and forensic accuracy. Experimental results show PromptShield boosts classification performance under attack conditions, achieving precision, recall, and F1 scores above 93%, while CIAF enhances ransomware detection accuracy in cloud logs using Likert-transformed performance features. Our integrated framework advances the automation, interpretability, and trustworthiness of cloud forensics and LLM-based systems, offering a scalable foundation for real-time, AI-driven incident response across diverse cloud infrastructures.
