How AI is Transforming the Future of Lending and Credit Risk Across Europe

From Brussels to Amsterdam, a clear picture emerged: European banks are at a pivotal moment in their AI journey, but the path forward varies dramatically by market.

From "If" to "When": The Acceleration of AI Adoption 

The shift in conversation has been striking. "Last year, when we had this discussion, the thinking was: it would be great if AI could…" Tim Mussche, Area Vice President EMEA at nCino, told executives at a breakfast briefing in Amsterdam. "Now, it's no longer if AI, it's when. And the use of AI has been pushed all the way from the board table down to the different teams to execute on." 

Yet adoption patterns differ. In Belgium, Benoit Lafort, Regional Vice President at nCino, outlined a more cautious landscape. While AI-based tools like CoPilot are widespread for communication and presentations, AI is rarely embedded in core lending decisions. "It's a buzzword," one Brussels participant noted. "There are often no proper resources or plans on how to integrate AI into the workflow." 

Current Use Cases: Where AI is Making an Impact 

Despite varying adoption speeds, clear use cases are emerging: 

• Administrative Efficiency: Banks are achieving quick wins through horizontal acceleration of daily tasks—email drafting, document preparation, and data input. Several institutions are now automating significant portions of credit approval memos, with one large bank reporting that most sections can now be written by AI, including quality control checks. 

• Data Analysis and Risk Assessment: Banks are harnessing AI to analyze large volumes of data, including supervisory statements and probability of default—though this remains limited by data availability. In the Nordics, banks are exploring real-time credit benchmarking and early warning systems that monitor customer behavior, sector trends, and market changes. 

• Compliance and Monitoring: Opportunities exist in verification, KYC data streamlining, and regulatory compliance. nCino has already developed tools where AI searches through applicable legislation, helping banks navigate complex regulatory demands. 

• Financial Literacy: In the Nordics, institutions use AI to educate clients on bank websites, explaining financial products and application processes—particularly valuable in markets where financial literacy varies. 

The SME Challenge and Opportunity 

Small and medium-sized business lending emerged as a key focus area. Participants see potential for AI to make credit access simpler and faster for SMEs, creating opportunities particularly for smaller banks. Prescoring techniques used in Denmark could be applied more broadly, though data availability remains a constraint—especially in Belgium compared to markets like Sweden, where salary information is publicly available. 

The Human Element: "AI Provides the Skeleton, We Add the Muscles" 

Despite AI's promise, executives across Benelux and the Nordics remain unanimous: humans must stay in the loop. 

"AI can provide the skeleton, but we have to put the muscles on it," one Amsterdam participant explained. Full automation of credit decisions is still seen as too risky and potentially harmful to customer relationships. The nuanced analysis required for complex, non-standard cases—particularly in emerging markets—still demands human judgment. 

Consumer preferences reinforce this. Participants noted that clients still prefer speaking with human beings for long-term or complex products like mortgages—and younger clients aren't necessarily more willing to engage with AI-driven services for these high-stakes decisions. 

The Transparency Challenge: Breaking the Black Box 

The most persistent concern across all markets? Explainability. 

"We see that AI can help in decision-making, but it needs to be explainable," said one Brussels participant. "If it's just a black box, that's not something that we, at our current stage, can accept." 

Banks want—and regulators expect—the ability to explain to customers why their loan was approved or rejected. Current AI models don't allow for this transparency. As one executive put it: there's a fear of "computer says no" without understanding why. 

The reliability of data compounds this challenge. "Very often, especially in the generation of AI, we don't quite know what we have created and where it has evolved," noted an Amsterdam participant. While AI can be free from human bias, executives recognized that humans create AI which leads to built-in bias. "Humans have bias, AI has hallucinations," one executive observed. 

Regulatory Uncertainty: A European Patchwork 

Regulation emerged as a significant hurdle, with different nuances across markets. Belgium's regulatory landscape is particularly unclear—the European AI Directive leaves room for interpretation, and the Belgian regulator has yet to apply specific guidelines. The Belgian market is also "maybe less acceptant of technological challenges," one participant noted. 

Current regulatory requirements stipulate that clients must be informed when AI makes decisions and have recourse to manual review. Participants expected that industry practice and regulatory adaptation would evolve together, though questions remain: Will regulation only allow improvements to current practices, rather than challenging assumptions like the need for humans in the loop? 

Technical and Talent Challenges 

Integration with legacy systems remains complex, with programming talent difficult to find. Some banks struggle to get AI to perform as desired. "95% of what we ask AI to do fails," one participant admitted. Good testing, mirroring, and installing checks can help, requiring a balance of trust and control. 

The skills required are evolving. As AI takes over more analysis, concerns emerge about new employees losing foundational analytical skills. The role of relationship managers will shift toward client relationships rather than document writing. New employees will need different capabilities: prompt engineering, the ability to explain how models work, and flexibility to adapt to rapidly changing technology. 

Looking Ahead: The Competitive Imperative 

Non-banking competitors are leveraging technology to offer faster, more streamlined services, forcing banks to accelerate AI adoption. Embedded finance and digital lending platforms are reshaping the landscape, especially in consumer finance. 

"In the long term, we are going to need to use AI to the full extent," said one Brussels participant. "We are one of the service providers that is really a good use-case for AI to replace us." 

An Amsterdam executive was even more direct: "The only way to survive this is to fully adopt it." 

Yet adoption will be nuanced. Client size and risk profile will influence AI’s role in the lending chain. For smaller clients, human judgment will be limited in the future due to cost. For larger clients pursuing riskier deals, the question of when AI can determine which clients need human oversight remains open. 
 
This is precisely the vision behind nCino's Digital Partners—role-based AI agents designed to amplify human capabilities rather than replace them. Working invisibly in the background, these agents create a "dual workforce" that allows banking professionals to focus on decision-making and relationship building while AI handles complex workflows and administrative tasks. 

With AI holding significant promise for automating routine tasks, improving efficiency, and supporting compliance, substantial challenges remain in integrating it into core credit decision-making. Explainability, human oversight, and regulatory alignment are essential to advancing AI adoption in European banking. 

The question is no longer whether banks will adopt AI—but how quickly they can do so responsibly. 

Ready to navigate the AI transformation with confidence? Discover how nCino’s Digital Partners can help European banks adopt AI —with full transparency and human oversight built in.