Artificial Intelligence Readiness for Compliance Risk Management: A Qualitative Case Study of a Vietnamese Maritime Logistics Firm

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Purpose – This study investigated how a maritime logistics firm in Vietnam could develop organizational readiness for the responsible use of artificial intelligence (AI) in compliance risk management. It addressed three connected questions: which compliance risks should be prioritized, what organizational conditions constrain AI adoption, and which implementation sequence is feasible for a resource-conscious firm. Design/methodology/approach – An applied qualitative single-case design was used at TRA-SAS. Evidence comprised three semi-structured interview sessions involving eight participants from leadership and key functional areas, internal process and business documents, public corporate reports for 2021–2025, and a 5 × 5 risk-assessment matrix. Thematic analysis was combined with a purpose-specific AI-readiness assessment covering data, technology, people, processes, finance, leadership, compliance culture, and security/legal governance. Findings – Eight material compliance risks were identified. Customs compliance (CR04) received the highest priority within the portfolio (likelihood 3, impact 5, score 15; High). The firm displayed an uneven M2–M3 digital-maturity profile: software and selected processes were near M3, while data standardization and system integration remained closer to M2. High-value AI opportunities were document checking, contract and obligation analysis, regulatory change monitoring, early-warning analytics, and compliance reporting. However, these use cases depended on data governance, standardized workflows, role clarity, human review, and model monitoring. Originality/value – The study connects compliance-risk materiality with purpose-specific AI readiness in an under-researched maritime logistics setting. It proposes a risk-first, readiness-gated, human-in-the-loop pathway that avoids treating AI adoption as a technology procurement decision.