As AI transforms banking, the challenge is to balance technological innovation with the human touch that builds trust. Sopra Steria Next’s Miguel Fernández Sanz offers a roadmap for achieving both.

Picture a traditional bank relationship manager preparing for a client meeting. Where once, this required scanning screeds of printed statements, handwritten notes, and disparate digital records to reconstruct a customer's financial profile. Today, that same manager has an AI assistant that instantly consolidates CRM data, call histories, and transaction patterns, delivering personalised insights before the customer even walks through the door.

This transformation from reactive banking to proactive customer centricity illustrates the profound shift underway throughout Europe’s financial sector. While 70% of consumers believe generative AI will positively impact banking within three years, the real question isn't whether AI will transform the industry, but how banks will manage this technological shift while maintaining the human dimension that underpins trust in financial services.

To understand this delicate balance, we spoke with Miguel Fernández Sanz, Financial Services Consulting Lead at Sopra Steria Next in Spain, who has witnessed banking's evolution from traditional to digital and now to AI-first operations.

AI adoption moves past efficiency

What sets today's AI revolution apart from previous technological waves in banking? For Fernández Sanz, the answer lies in its accessibility and purpose.

While generative AI is democratizing access to advanced technology, most banks are still in the exploratory or pilot phase, with large-scale, ROI-positive deployments remaining the exception rather than the rule.

But democratisation alone is not the sole driver of adoption. Banks now regard AI as more than a vehicle for operational efficiency – it is a strategic enabler of customer centricity. "Efficiency is the catalyst, but personalisation is the objective," notes Fernández Sanz. This shift from cost reduction to customer value creation marks a fundamental change in how financial institutions approach AI investment.

The numbers support this strategic pivot. According to Sopra Steria's Digital Banking Experience Report 2025, 62% of banking decision-makers rank improving risk management as AI's highest impact area, while 40% cite high-impact potential and only 22% express critical concerns. This measured optimism suggests banks are moving beyond experimental phases towards strategic implementation.

From predictive to generative: redefining customer interactions

The distinction between predictive and generative AI reveals the scope of transformation ahead. Traditional predictive AI has long powered credit scoring, fraud detection, and anti-money laundering systems, optimising existing processes through data analysis.

Generative AI represents something more revolutionary. "It enables the creation of new content, from text to code," explains Fernández Sanz. "More importantly, it allows for hyper-personalisation: tailoring interactions and products to each customer's specific situation."

Most generative AI use cases in banking currently focus on internal efficiency and employee support. Customer-facing applications, such as advanced conversational agents, are still largely in pilot phases due to concerns around accuracy, bias, and customer satisfaction "We're currently working with several banks to make their conversational agents more natural and human-like," notes Fernández Sanz. "The goal is to move from scripted exchanges to genuine dialogue between banks and clients."

Document management provides another concrete example. Documentation tasks that previously required significant manual effort can now be streamlined through generative AI, especially in areas where documentation remains fragmented or unstructured.

Trust through transparency and value

Despite AI's potential, consumer trust remains a critical hurdle. Only 27% of consumers currently trust AI for financial advice, highlighting the challenge banks face in bridging technological capability with customer confidence.

Fernández Sanz believes the solution lies in demonstrating value rather than concealing AI's role. "If a bank uses AI to proactively offer the right product at the right time – say, a credit offer just when a customer is planning a holiday – that's valuable and convenient. The customer feels understood rather than targeted."

Transparency is essential in this trust-building process. "If a relationship manager recommends a product, the customer should understand why: what data or insights support that recommendation," he emphasises. This approach transforms AI from an opaque mechanism into a tool for enhanced customer understanding.

While neobanks have set new standards in anticipating customer needs and clear communication, traditional banks still enjoy higher trust levels. However, both must focus on transparency in AI-driven processes to maintain and strengthen customer confidence.

Navigating the regulatory landscape

Europe’s regulatory landscape—shaped by the GDPR and, looking ahead, the forthcoming Financial Data Access Regulation (FIDA) and the new AI Act—presents both challenges and opportunities for AI adoption in banking. Rather than viewing regulation as a constraint, Fernández Sanz sees it as a framework for responsible innovation.

"Banks already follow strict compliance processes: algorithms used for credit scoring must be documented, explainable, and auditable by regulators," he notes. "The same approach must apply to AI systems."

This regulatory foundation strengthens, rather than inhibits, customer trust. When banks can demonstrate that their AI systems operate within clearly defined compliance frameworks, they reassure customers that innovation does not come at the expense of fairness or security.

Internal adoption: empowering not replacing employees

Contrary to fears of AI replacing human workers, Fernández Sanz observes enthusiasm among bank employees.”. While there is growing interest and informal adoption of GenAI tools among employees, banks must address concerns around data security, compliance, and the evolving nature of job roles through robust governance and change management”.

The challenge is not adoption but governance – ensuring these tools are used securely and in compliance with internal data policies. Success requires clear AI governance and change management strategies. "Governance ensures compliance and defines which use cases to prioritise. Change management helps employees learn how to use AI effectively in their daily tasks."

Internal AI applications are already improving efficiency. Chatbots consolidate CRM data, helping relationship managers better understand clients before interactions. AI assists with email writing, document summarisation, and presentation generation, enhancing rather than replacing human capabilities.

The future: agentic AI and human oversight

Looking ahead, autonomous AI agents represent the next frontier, though with important limitations. " Agentic AI, or fully autonomous decision-making systems, remains a long-term vision. In the short to medium term, human oversight is essential, particularly in critical areas such as credit approval and compliance." explains Fernández Sanz.

These systems will likely prove valuable in operational processes like document handling and fraud detection. However, critical domains like credit approval and compliance will continue requiring human oversight. "As I like to say, AI can drive the car, but someone must still hold the wheel."

This human-in-the-loop model reflects a broader philosophy about AI's role in banking: augmentation not replacement of human judgement. The most successful implementations will combine AI's analytical power with human judgment, creating experiences that are both technologically sophisticated and authentically human.

Strategic priorities for banks

For banks developing AI strategies, Fernández Sanz emphasises three core imperatives:

Strategic governance and compliance: Develop robust AI governance frameworks aligned with regulation, ensuring transparency and customer trust through collaboration between IT and business units.

Human-centric innovation: Adopt an "AI-first" mindset by embedding AI at the heart of customer interactions while maintaining human oversight and ethical standards.

Scalable infrastructure: Build cloud-native, API-driven platforms capable of securely supporting multiple AI use cases at scale, ensuring both resilience and regulatory alignment for both current and future applications.

Spanish banks are already moving in this direction, building AI-first architectures that support rapid experimentation and compliant deployment. This infrastructure-led approach allows institutions to scale innovations quickly while preserving trust and operational integrity.

Toward democratised financial guidance

The true promise of AI in banking lies not in efficiency alone, but in its potential to democratise access to personalised financial advice. "In the past, only private or wealth management clients had access to personalised financial advice," notes Fernández Sanz. " AI has the potential to democratize access to personalized financial advice, but realizing this vision will require overcoming significant challenges in data quality, regulatory compliance, and model maturity."

This shift represents one of AI's most significant potential impacts: transforming banks from transaction processors to financial wellness coaches. By providing personalised guidance at scale, AI can improve financial well-being across all customer segments, not just the affluent.

Risks such as algorithmic bias, hallucinations, and data privacy concerns are key factors driving a cautious and incremental approach to GenAI adoption in banking.

Banks that succeed in this transformation will be those that view AI not as a substitute for human connection, but as its amplifier. They will use technology to deepen their understanding of customer needs, to respond more quickly and accurately, and to deliver advice and products that are more genuinely relevant.

This is the future that Fernández Sanz envisions, one banks must actively orchestrate: a harmonious blend of technological capability and human insight that makes banking more inclusive, efficient, and genuinely responsive.