Photo by Jessica/Adobe Stock
Artificial intelligence (AI) is becoming part of the digital systems that help global supply chains move goods, share information and manage disruption. In transportation, logistics, warehousing and inventory management, AI is increasingly used to forecast demand, optimise routes, process documents, monitor assets and support warehouse operations. These applications may improve efficiency and responsiveness, but they also create new forms of dependence on software, data, vendors and shared digital infrastructure.
This matters because supply chains are already highly interconnected. If, for example, many firms rely on the same AI tools, cloud services or software providers, a single failure could affect multiple organisations at once. That raises new questions about resilience, accountability and insurance. Insurers, operators and regulators need to understand not only how AI is being adopted, but also how AI-related failures could spread across firms, disrupt operations and generate losses across different types of insurance cover.
Our research examined how growing reliance on AI in supply chains may affect operational resilience and insurance risk. We focused on transportation, logistics, warehousing and inventory management, and explored where AI is being adopted, how it may fail, and what kinds of losses or disruption could follow.
The study combined several methods. It reviewed academic and industry literature on AI use in supply chain operations, analysed real-world AI-related incidents and lawsuits, and conducted stakeholder interviews with experts from insurance, law, technology, logistics, warehousing and ports. It also developed a set of scenarios to test how AI-related failures could create correlated losses across multiple firms and insurance lines. Together, these methods were used to identify key risk pathways, assess how adoption may evolve over the next five years, and develop recommendations for operators, insurers and regulators.
The research found that AI adoption in supply chains is increasing, but unevenly. Lower-risk uses such as monitoring, forecasting, route optimisation and document processing are spreading more quickly than more autonomous forms of execution. In many cases, AI is being introduced through software upgrades and digital platforms, meaning that adoption may grow quietly and sometimes without firms clearly recognising where AI is shaping operational decisions.
The study also found that AI-related failures may create new forms of hidden dependency and insurance risk. Firms that appear independent may rely on the same software provider, cloud platform or AI-enabled service. This means that a single weakness, outage, cyberattack or regulatory change could affect multiple organisations at once and generate losses across several insurance lines. At the same time, governance and insurance responses remain unsettled. Responsibility for AI-related failures is often difficult to allocate across operators, developers, vendors and integrators, while insurers are still adapting their underwriting, exclusions and coverage approaches to reflect emerging AI-related risks.
Supply chain operators and technology providers can reduce risk by putting stronger controls around how AI is used in live operations. This includes understanding where AI is embedded in workflows, keeping meaningful human oversight over high-impact decisions, monitoring systems for abnormal behaviour and maintaining fallback options if systems fail.
Insurers can improve their response by looking more closely at how AI is used in the businesses they cover, including where firms depend on shared vendors, models or digital services. Regulators can help by clarifying governance expectations, improving transparency and considering whether existing insurance arrangements are sufficient for extreme AI-related losses that could affect many firms at the same time.
