On the final day of Sibos in Miami, one of the most critical discussions of the conference tackled how data policy, artificial intelligence, and evolving digital infrastructure are reshaping global supply chains and trade finance.
The session was chaired by Julia Streets MBE, founder and Chair of Streets Consulting, who opened the debate by emphasising that as international commerce faces geopolitical shifts and technological convergence, getting data flows right is the prerequisite for cross-border trade resilience.
Joining Julia on stage was an international panel representing the full spectrum of global trade and technology: Emmanuelle Petele (Deputy Head of Trade Services, France, Société Générale), Makiko Toyoda (Global Head of Trade and Supply Chain Finance, International Finance Corporation / IFC), Natasha Condon (Global Head of Trade and Working Capital Commercialisation, J.P. Morgan), and Vinay Lawrence Mendonca (Head of Product, Trade and Supply Finance and Corporate Channels, Finastra).
The discussion laid out a clear blueprint for the future of trade finance, centring on five strategic themes at the intersection of technology, regulation, and human expertise.
Cross-border data ownership, nation-state security concerns, and localised storage rules have moved to the centre of corporate strategy. Operating across multiple jurisdictions requires institutions to evaluate local regulations, onshore storage mandates, and data privacy rules before deploying cloud-native or AI-powered solutions. In addition, implementing advanced AI capabilities quickly reveals that automated tools are only as effective as the underlying data fed into them.
Internally, banks and corporates must prioritise structuring data across invoices, amendments, fees, maritime documents, and liquidity positions. Crucially, panellists agreed that consolidating global trade data into a single, centralised repository is no longer viable. Instead, the industry is transitioning toward distributed data architectures connected by event-driven streaming (such as Kafka).
Under this model, data remains at its native source while status updates, such as document presentation rejections, are published instantly to all relevant stakeholders. To enable seamless communication across distributed platforms, institutions must adopt semantic canonical terms, such as those established by the International Chamber of Commerce (ICC) Digital Standards Initiative, ensuring disparate systems interpret trade events identically.
A profound digital divide currently separates advanced economies from emerging markets. While approximately 90% of firms in developed nations utilise AI for trade activities, only 30% of firms in low-income countries currently have access to AI tools. Without targeted intervention, this technological gap threatens to widen the existing $2.5 trillion global trade finance gap.
To address this disparity, international initiatives such as the World Bank Group’s Digital Trade for Africa program are establishing foundational digital infrastructure across pilot countries, including Benin, Côte d’Ivoire, Ghana, Nigeria, Kenya, and Rwanda, focusing on connectivity, electronic transactions, e-signatures, and cross-border data frameworks. Statistical data indicates that improving digital infrastructure by 50% in emerging economies can increase national income by 15%. Real-time transactional data and automated processing democratise access to international commerce, allowing small and medium-sized enterprises (SMEs), smallholder farmers, and deep-tier suppliers (Tier 1 through Tier 3) to secure trade credit without submitting traditional audited financial statements.
While regulatory progress has delivered electronic-document reforms across 60 jurisdictions and established 40 electronic registries, universal adoption of the Model Law on Electronic Transferable Records (MLETR) and cross-border legal interoperability remain complex challenges. In multi-jurisdictional transactions, one leg of a trade may sit under contract law while another operates under local MLETR legislation, requiring precise legal and technical interpretation.
What’s more, financial institutions operating under frameworks like the EU AI Act must adopt risk-based documentation standards when deploying automated algorithms. As artificial intelligence evolves from answering routine queries to executing autonomous trade instructions via agentic AI, building robust trust and control frameworks becomes paramount. When an autonomous agent instructs a bank, the institution must verify the agent’s identity, delegated authority, and security parameters with the same legal rigour applied to human signatures. Panellists emphasised that universal adoption of the Legal Entity Identifier (LEI) and verifiable organisational identities is vital to preventing cross-border fraud and protecting supply chain integrity.
Financial institutions are already deploying applied AI solutions to eliminate operational friction and accelerate execution. For example, automated guarantee text vetting tools can instantly validate client commercial contracts against ISP98 rules and internal bank parameters in seconds, replacing hours of manual review. Similarly, credit application compiling, which merges client financial statements with account manager call reports, can now be executed in seconds, dramatically reducing turnaround times for three-to-six-month working capital facilities.
In risk management, AI transforms credit assessment from a static, annual review to continuous real-time risk monitoring. Rather than relying on outdated financial statements, automated systems continuously evaluate live signals, including payment delays, invoice dilution, cash flow changes, and supplier term adjustments. This granular visibility allows lenders to price risk more accurately and expand credit capacity to previously underserved businesses. Furthermore, combining AI data connectivity with tokenisation and stablecoins presents immense opportunities to embed liquidity directly into commercial venues and extend financing to deep-tier suppliers.
Automating formulaic document checking removes repetitive administrative tasks, allowing senior trade operations specialists to focus on high-level risk analysis, complex structuring, and fraud detection. Rather than replacing human expertise, institutions are consolidating specialist talent into regional expert hubs to maintain technical competency across global booking locations. Correspondent banking networks and human relationships remain irreplaceable for interpreting geopolitical events, country risk, and volatile market disruptions that raw data alone cannot capture.
At the same time, AI democratises technology across front-office sales and relationship management teams. Without software engineering backgrounds, business users can now prototype bespoke solutions, such as “vibe coding” guarantee wording tools, to address client pain points directly. Relationship managers can synthesise public and private market data in real time, delivering highly tailored working capital proposals and competitive benchmarking to corporate clients at scale.
To successfully navigate the transition toward AI-driven digital trade, the panel called on market participants to execute across three core priorities:
