By: Robert Downs
Corporate lending is at an inflection point, with market forces reshaping the landscape at an unprecedented pace. In Europe, corporate loans are growing at 6% annually and syndicated lending markets are expanding three times faster – reaching 15-18% annual growth[1]. Yet beneath these promising figures lies a fundamental challenge: most banks’ infrastructure was never designed to handle this volume or complexity.
Client expectations have evolved dramatically, demanding greater accuracy, transparency, and speed from their lending partners. Meanwhile, the reality inside many institutions remains starkly different, with processes still heavily reliant on emails, spreadsheets, and manual steps that simply cannot scale. Regulatory pressure compounds these challenges, requiring clean data, robust controls, and end-to-end transparency that manual workflows struggle to deliver.
Though it is as clear as ever that modernisation is no longer optional. The use of AI, automation, and tokenisation is now transforming how banks originate, manage, and service loans, creating pathways to excellence that were previously unimaginable.
AI, automation, and data-driven efficiencies
In conversations around AI in lending, banks are no longer just asking “why AI?” but are now looking a step further and asking instead, “what sequence should we implement it in?” The growing industry consensus, it would then seem, is that AI adoption is inevitable. The only question is how quickly institutions can move from pilot projects to production-ready systems.
Yet a significant competency divide is already taking root. Banks that have implemented AI-driven workflows are gaining the potential to close syndicated deals in just two to three days, while institutions relying on manual processes require up to three weeks for similar transactions. This performance gap reveals itself most clearly in data extraction and validation. A typical syndicated lending deal involves extracting and validating 40 to 50 distinct data points – including rates, covenants, conditions, and waterfall structures – yet most banks today still manually mine, validate, and reconcile this information before it reaches their core systems[2].
Leading institutions are deploying AI across three applications. First, they’re using it to read unstructured data from emails, PDFs, and documents, dramatically reducing manual work while improving data quality. Second, they’re automating checks to detect inconsistencies and exceptions, allowing teams to focus on high-value decision-making rather than routine validation. Third, they’re eliminating low-value manual tasks entirely, creating capacity for more strategic client service and risk management.
However, successful deployment requires thinking beyond point solutions. Banks must implement AI as production-ready systems with validated responses, auditability, and structured escalations to humans when needed. The goal isn’t to replace human judgment but to automate simple, structured processes so decisions flow seamlessly into systems of record.
Several barriers continue to slow adoption. Fragmented data silos across departments and markets create additional friction, especially in institutions operating across multiple European jurisdictions. Institutions, then, need to remain agile to quickly integrate new AI capabilities, as regulations like Europe’s AI Act add complexity.
Ultimately, the success of any AI initiative depends on data quality. You can’t tokenise what you can’t trust and can’t automate what’s not structured. It all starts with data. Banks must prioritise data standardisation and cleansing as the foundation, ensuring information is accurate, consistent, and structured before layering on automation and intelligence. Without this groundwork, even the most sophisticated AI tools will struggle to deliver meaningful results.
Tokenisation and new models of collaboration in syndicated lending
While AI addresses immediate challenges, tokenisation represents a longer-term transformation. Just a year ago, tokenisation was considered far off in the future. Yet it is quickly becoming an enabler of both standardisation and interoperability. At its core, tokenisation promises to create a common language across the lending ecosystem – ensuring everyone discusses the same instruments using identical formats and structures.
Early applications focus on wrapping loans as tokens that represent ownership of financial benefits, rather than tokenising the complete loan structure itself. However, the spectrum of use cases is expanding quickly, from single-line loans to complex collateralised loan obligations (CLOs) settling on blockchain infrastructure. These developments can deliver tangible benefits in the form of faster settlement times, reduced counterparty risk, and broader distribution channels for credit instruments.
Yet, as with many such innovations, significant hurdles remain. Without network interoperability and shared industry standards, tokenisation cannot reach its full potential. The same fundamental principle applies here as it does with AI: data quality and integrity must come first.
This reality is driving new collaboration models that bring together banks, technology providers, and market operators through infrastructure partnerships. Rather than forcing banks to replace core systems, these partnerships layer intelligence and connectivity on top of existing platforms. Banks maintain their systems of record, such as Loan IQ, which serves as the hub and golden source for loan data, while gaining seamless integration with external networks. This approach reduces vendor complexity, delivering complete workflows through single integration points rather than requiring banks to manage multiple disparate systems.
Open APIs play a crucial role in these partnerships, yet they are still underdeveloped in corporate banking. Expanding API frameworks allows each platform to specialise in its strengths while maintaining consistent data quality across internal processes. By positioning core systems as hubs that manage data and ensure consistency, banks can integrate externally without the risk of “big bang” transformations that disrupt operations.
Streamlining workflows, cutting costs, and driving growth
The path forward requires both vision and action. Five priorities that can accelerate modernisation while managing risk effectively include:
· Loan servicing automation – given the growth in syndicated lending, automating workflows is the only viable path to maintaining service quality without proportionally scaling headcount.
· Modular, API-driven platforms – banks need infrastructure that allows them to upgrade quickly to deliver new capabilities, so that AI features, tokenisation protocols, or external integrations can be deployed rapidly as market conditions evolve.
· Data readiness enterprise-wide – only comprehensive data standardisation will unlock true ROI from transformation initiatives. Banks cannot afford to modernise in isolated pockets while leaving core data fragmented and unstructured.
· Collaboration and education – with significant expertise retiring over the next decade, institutions should invest in training for the next generation coming through to ensure there is a continuity of specialised knowledge. Industry-wide data utilities and expanded open API frameworks can create shared standards that benefit all participants.
· Phased transformation – while technology readiness may tempt ambitious overhauls, successful modernisation requires balancing innovation with stability. Sustainable change encompasses people, processes, and systems, all working in harmony.
[1] Eric Li, Head of Global Banking Research, Coalition Greenwich, Finastra IFT Summit, 31 March 2026
[2] Vihang Patel, Head of Data and AI, Marketnode, Finastra IFT Summit, 31 March 2026
