Financial firms struggle to fully scale AI amid data silos & security

Most financial services providers worldwide are adopting artificial intelligence (AI), but nearly half remain stuck between early experimentation and fully integrated AI operations, according to new international research.

The findings highlight key barriers and strategic requirements for achieving scale with AI, underscoring the importance of unified data governance and hybrid deployment models.

Hybrid AI models

The study, based on a survey of 155 executives from financial institutions across several continents, found that 62% of respondents are using a hybrid approach to AI.

This involves deploying AI across combinations of public cloud, data centres, and edge environments, enabling firms to access and process data wherever it resides. The research suggests this strategy is now essential for managing legacy systems and efficiently scaling AI initiatives.

Ninety-one percent of participants rated a hybrid AI approach as highly valuable, and 97% indicated they are already using at least one AI or machine learning use case in their businesses. Despite near-universal take-up, the benefits of AI risk being limited by the way organisations use and govern their data.

Security and silos

Data security emerged as the top barrier preventing AI from being adopted at scale. Nearly all (97%) of financial services organisations in the survey reported that data silos are holding back the development and deployment of effective AI models.

The persistence of these silos, along with varying approaches to data governance, is impeding full-scale transformation.

Although firms are ambitious in their AI efforts, only 26% reported achieving full, enterprise-wide adoption. Nearly half (48%) said they have progressed beyond basic experimentation but have not yet fully integrated AI across their operations.

Unified governance

The importance of consistent governance features prominently in the results. Eighty-four percent of surveyed organisations said a unified approach to data governance and security across all technological environments is either 'critical' or 'very important'. This reflects a broader industry focus on regulatory compliance, privacy, and data protection, particularly as sensitive financial data is increasingly leveraged for AI development and decision-making.

"The report confirms that a data-anywhere, hybrid strategy is non-negotiable. It also highlights that infrastructure alone isn't enough. To truly bridge the implementation gap, financial institutions need a unified data and AI platform that ensures consistent governance and security across all environments. This is the only way to build trust, manage risk, and accelerate AI adoption at scale," said Adrien Chenallier, Global Director, AI Solutions for Financial Services at Cloudera.

Investment priorities

Financial services firms are prioritising security in their AI investment strategies. Twenty-five percent of respondents cited security as their main criterion when selecting vendors for AI platforms. This renewed attention to security and governance aligns with broader trends in digital transformation, where data sovereignty and privacy are frequently highlighted by both regulators and industry bodies.

The importance of trust, compliance, and robust governance frameworks is echoed by industry leaders as they look to advance further in their AI journey. Firms face increasing pressure to manage risk effectively and remain competitive in a rapidly changing digital landscape.

"For our members, the reality is clear: AI can only deliver its full potential when data sovereignty, privacy, and trust are guaranteed. Our survey results highlight both progress and gaps, underscoring that AI success will depend not just on the scale of investment, but on strategic decisions around infrastructure, vendor partnerships, and robust data governance. We're pleased to share these insights in partnership with Cloudera, helping our global members navigate this rapidly evolving landscape," said Gary Wright, Managing Director, Finextra Research.