Endava: Behind the AI Adoption Gap in Financial Services

Financial institutions are confident about adopting agentic AI. They’re just not paying for it yet.

Endava surveyed 1,000 senior leaders across nine countries and found that 92% believe their organisations are ready to deploy autonomous AI systems. Just 36% have allocated budgets to make it happen.

What stands between ambition and implementation is not conviction. Leaders understand the technology and its applications. What’s missing is the infrastructure, governance and technical capability to run these systems at scale.

Agile processes create bottlenecks

Three-quarters of surveyed leaders said agile methodologies are slowing them down. Frameworks designed to speed up software development now struggle to accommodate AI systems that operate on different timescales.

Regulatory complexity and ecosystem integrations are where the friction occurs. Both were cited by 49% of respondents as areas where agile processes fail to keep pace. An AI system connecting with multiple third-party platforms while navigating cross-border regulations and compliance requirements doesn't fit neatly into two-week sprints.

Agile still has its place. Some 86% see value in these methods for certain tasks. The question is when to use them and when to work around them. Most leaders, 76%, expect their organisations will need AI-native operating models within two to three years. Nearly all surveyed (94%) think this shift will determine competitive standing.

Few have made the transition, however. Only 16% of organisations describe themselves as AI-native. Italy leads at 25%, followed by the United States at 24%. France sits at 6% and the UAE at 4%. The UK matches the global average of 16%.

Agentic AI expected to unlock market potential

More than 80% of surveyed leaders expect agentic AI to open new markets or create different revenue streams. The immediate applications centre on fraud detection, financial crime prevention and operational continuity.

Banks lose billions annually to fraud. Regulators continue tightening anti-money laundering and know-your-customer requirements. System failures, however brief, trigger penalties and erode customer trust. Agentic AI can process transactions and identify anomalies faster than human teams.

The longer view involves continuous operation. Leaders surveyed anticipate 24/7 service, faster product launches and more personalised customer interactions. Some expect autonomous systems to handle routine tasks without human involvement, redirecting staff to work requiring human judgement.

Part of this is defensive. Digital banks and fintech firms built their operations around speed and user experience. Traditional institutions carrying legacy systems and deeper regulatory obligations are working to close that gap.

Data privacy and regulatory uncertainty top concerns for financial leaders

Data privacy, regulatory uncertainty and transparency in decision-making came up most frequently when leaders discussed concerns about agentic AI deployment.

When an AI system makes a lending decision, blocks a transaction or flags an account, customers and regulators want explanations. Explainability becomes harder when systems learn and adapt autonomously. Data privacy grows more complicated when AI models require access to information across jurisdictions with different rules.

Nearly half of surveyed organisations, 47%, are building ethical guidelines into AI development. The same percentage are implementing transparency and explainability measures. Another 46% are strengthening data privacy protections, while 44% are establishing governance frameworks for AI use.

Some 42% of firms are working to align agentic AI with regulatory requirements. Just over a third (37%) are training employees on responsible AI use. The figures indicate that organisations recognise governance failures would convert the technology from asset to liability.

Matt Cloke, CTO at Endava, says: “Agentic AI represents a step-change in how financial services organisations operate and innovate. The opportunity is clear, but so is the responsibility. Our research shows that those who build AI-native operating models, backed by strong governance, will be the ones to lead the next era of financial services.

“At Endava, we’re already adopting this approach with Dava.Flow, our AI-enabled engagement lifecycle. We know that success lies in adapting quickly, embracing multidisciplinary teams and balancing innovation with organisational health.”