As software systems grow in complexity, maintaining traceability—the ability to establish and manage relationships between software artifacts—becomes increasingly critical for supporting requirements validation, change impact analysis, and compliance assessment. Despite its recognized importance, the manual effort required to create and maintain trace links presents a significant barrier to adoption and limits traceability’s practical value. This research investigates how Large Language Models (LLMs) can address fundamental traceability challenges across four key areas: improving trace-link prediction in data-scarce environments, generating software artifacts and establishing traceability in projects lacking formal documentation, tailoring automated traceability to support accountability in regulated domains such as Cyber-Physical Systems, and enhancing software maintenance through traceability-supported differential testing workflows. By addressing these challenges, this research aims to make comprehensive traceability accessible across all software projects, enabling existing engineering practices to more effectively leverage trace relationships for enhanced maintainability, safety, and regulatory compliance.