By Smitha Shetty, Regional Director, APAC, Achilles Information Limited
In boardrooms across the Asia-Pacific region, conversations around supply chain due diligence have reached a point of maturity. The question is no longer whether organisations should strengthen oversight, but how they can do so credibly across supply chains that stretch across tens of thousands of suppliers, multiple jurisdictions and increasingly complex regulatory expectations. Within these discussions, artificial intelligence is rarely positioned as a silver bullet. Instead, it is increasingly recognised as a pragmatic response to a scale challenge that traditional, human-led approaches were never designed to manage.
This evolution is particularly evident in sectors such as energy, maritime, utilities and manufacturing, where supplier ecosystems are vast, deeply interconnected and operationally critical. What consistently emerges from conversations with procurement, sustainability and risk leaders is not resistance to due diligence itself, but frustration with its operational limitations. Manual reviews, periodic audits and spreadsheet-driven assessments can deliver depth in isolated cases, yet they struggle to keep pace with the volume, velocity and diversity of data now required to demonstrate meaningful and defensible oversight.
Leaders managing complex global value chains face a shared reality: traditional third-party assessment models allow only a small fraction of suppliers to be reviewed in any given year. The real value of AI lies not in replacing existing controls, but in fundamentally reshaping the due diligence workflow. By pre-processing vast volumes of supplier data, AI enables prioritisation at scale. Information can be scanned, structured and assessed automatically, allowing teams to focus their expertise on exceptions and higher-risk cases rather than expending effort on routine reviews. Human judgement remains central, but it is applied more deliberately—where context, experience and engagement matter most.
This shift is reflected in the practical experience of Achilles and the application of AI across global supply chains. Through AchillesAI, machine learning models are deployed to automate early-stage checks across supplier submissions, significantly improving both speed and consistency. By extracting key data from policies and procedures, AchillesAI auto-populates disclosure templates with ready-to-submit facts and narrative responses. The result is reduced administrative burden, improved data accuracy and greater confidence in meeting complex regulatory and customer requirements.
One of the most significant advantages of AI-enabled automation is anomaly detection. Advanced models can identify illogical combinations of declarations, inconsistencies between stated policies and supporting evidence, or unusual patterns in emissions, workforce or safety data. These signals are often extremely difficult to detect through manual review, particularly at scale. By surfacing them early, organisations can prevent superficial or duplicated responses from passing through unchecked and strengthen the reliability of data used for reporting, assurance and decision-making. Crucially, these insights are explainable, allowing teams to understand why a supplier has been flagged and determine the appropriate response.
Unstructured data remains one of the most persistent challenges in supplier due diligence. Information rarely arrives in standardised formats. It is often submitted as PDFs, scanned certificates, invoices or free-text explanations generated across diverse geographies and levels of digital maturity. AchillesAI addresses this challenge by reading, classifying and extracting supplier data at scale, transforming thousands of unstructured documents into structured, auditable insights in near real time. This capability also creates a stronger and more transparent audit trail—an increasingly important requirement as regulatory scrutiny intensifies worldwide.
AI-enabled automation also has a tangible impact on the supplier experience, particularly for micro, small and medium enterprises. Smaller suppliers are frequently required to complete multiple questionnaires for different customers, many of which overlap in content but differ in format. This creates disclosure fatigue and can discourage meaningful engagement. Centralised data repositories, reusable disclosures and greater standardisation reduce duplication and make participation more manageable, while still preserving the depth and integrity of information required. In doing so, suppliers are better supported in meeting complex regulatory expectations with confidence.
Looking ahead, industry leaders expect AI to move further upstream in the due diligence process. The emerging vision is one in which many checks are performed at the point of submission, enabling organisations to manage global supply chains with greater efficiency and focus. Real-time alerts can highlight emerging risks as data changes, rather than relying solely on periodic reviews. More refined supplier segmentation can distinguish between those that require support and capacity building and those that warrant closer monitoring. Predictive risk scoring will increasingly guide where limited human resources should be deployed.
From these discussions, a clear theme emerges. AI is not removing responsibility from organisations—it is strengthening their ability to exercise it with greater confidence, consistency and scale. The result is a hybrid governance model in which AI enhances efficiency and detection capability, while humans retain responsibility for interpretation, engagement and accountability. In this sense, AI does not substitute governance; it provides the infrastructure that makes credible governance possible. As supply chains continue to grow in size and complexity, AI’s role is shifting from experimentation to enablement—supporting due diligence that is robust, defensible and fit for an increasingly demanding regulatory and stakeholder environment.
