Financial services are betting big on AI, but most projects remain in pilot mode, study finds

Written by Pooja Sharma

Approximately 92% of financial services industry decision makers agree that improving data quality is critical to AI success

Financial services organizations are leading the way with strong ROI on AIOps initiatives, but only 12% of AI projects are fully deployed across the enterprise, with a significant 62% still in the pilot or development stages.

Riverbed, an observability AIOps company, has released financial services industry findings from a global survey. ‘The future of IT operations in the AI ​​era,‘ It examines the level of AI readiness across the financial services sector. The results highlight the growing implementation gap as organizations move from AI ambitions to real-world impact. Nearly all financial services decision makers (92%) agree that improving data quality is critical to AI success, but progress remains uneven. Only 12% of AI initiatives have achieved full enterprise-wide adoption, and a significant portion of 62% are still in the pilot or development stages, highlighting the challenges of operationalizing AI in one of the world’s most regulated and risk-sensitive industries.

However, the financial services sector continues to demonstrate strong confidence in the value of AI and AIOps, with 89% of organizations reporting that the ROI from their AIOps investments is higher than expected, reinforcing the industry’s reputation for disciplined, value-driven technology adoption. Additionally, nearly two-thirds (62%) of respondents expressed high confidence in their company’s AI strategy. However, despite this optimism, financial services organizations continue to be affected by gaps in AI adoption. With increasing pressure to optimize operations, strengthen compliance, reduce risk, and deliver superior digital experiences, the industry is increasingly constrained by data readiness, operational complexity, and the ability to scale AI beyond pilot initiatives.

“Financial services institutions are among the most sophisticated and disciplined adopters of AI, and our research shows they are already reaping significant benefits,” he said. Jim Gargan, Chief Marketing Officer; At Riverbed. “However, the sector operates under unique pressures, including intense regulatory oversight, zero tolerance for downtime, and a critical need for data accuracy. What is clear is that success now depends on IT simplification, observability tools and vendor consolidation, improved data quality, and OpenTelemetr “At Riverbed, we are actively supporting some of the world’s largest financial services organizations to bridge this gap and transform.” Translating AI ambitions into real-world operations. ”

AI ambition meets operational reality

For financial services institutions, success with AI is not just about experimentation. It depends on operational readiness. According to the study, only 40% of financial services organizations feel fully prepared to operationalize an AI strategy today. Data remains the most important constraint, with only 43% fully confident in the accuracy and completeness of all their organization’s data, the lowest level of confidence across all industries surveyed.

Importantly, the industry understands what is at stake. 92% of financial services respondents agree that improving data quality is critical to AI success, the highest percentage of any industry. This reflects a deep recognition that without reliable, high-quality data, it is difficult for AI initiatives to move from proof of concept to production.

Operational complexity drives simplification

These data challenges are further exacerbated by the complexity of today’s IT environments. To support digital services, real-time transactions, and growing AI workloads, financial services organizations have amassed fragmented toolsets that limit visibility and slow decision-making. Today, on average, IT teams use 13 observability tools from nine different vendors, creating blind spots across applications, networks, and user experiences.

As a result, 96% of organizations in this space are actively integrating tools and vendors across their IT operations, and 95% agree that a unified observability platform makes it easier to identify and resolve operational issues. Remarkably, 95% are considering new vendors as part of this consolidation, the highest level of all industries surveyed, demonstrating a willingness to rethink long-standing technology relationships in favor of platforms that can reduce risk, improve integration, and support AI at scale.

Unified communications performance becomes business critical

As financial services continue to digitize customer engagement and internal workflows, the performance of unified communications (UC) tools has become business-critical. Employees now spend 41% of their workweek using UC tools, and nearly two-thirds say UC tools are essential to effectively performing their jobs. However, performance is still unstable. Only 47% of financial services organizations are very satisfied with the performance of their UC, and 44% regularly report issues with video calls, messaging platforms, and collaborative workspaces.

These challenges create significant operational constraints. UC-related issues account for 16% of all IT tickets and take an average of 41 minutes to resolve, with nearly 1 in 5 tickets taking more than an hour. In a space where responsiveness and availability directly impact customer trust, limited visibility and high support demands continue to hinder productivity and experience.

OpenTelemetry powers observability at scale

To overcome fragmented visibility and support AI-driven operations, financial services organizations are increasingly turning to open, standardized observability frameworks. OpenTelemetry plays a critical role by enabling consistent data collection and correlation across applications, infrastructure, and user experience. This is a prerequisite for trustworthy AI in complex and regulated environments.

Encouragingly, the study shows that financial services organizations are leading the way across the board in OpenTelemetry adoption, with 92% already leveraging the framework. Nearly all respondents (96%) say cross-domain correlation is important to their observability strategy, while 99% agree that OpenTelemetry reduces vendor lock-in and increases flexibility. Importantly, 97% see this as a foundation for future initiatives such as AI-driven automation, reinforcing its role as an enabler of long-term AI scalability.

AI data movement and network performance are central

As AI efforts mature, attention is shifting from the models to the data movement that powers them. Financial services organizations value moving AI data more than any other sector surveyed, with 94% saying it is critical to their overall AI strategy and 37% saying it is important and fundamental to how they design and execute AI.

As AI data is increasingly distributed across public cloud, edge, and colocation environments, network performance and security have emerged as critical success factors, with 81% of respondents saying they are essential, the highest among industries. Looking ahead, 76% of financial services organizations plan to have an AI data repository strategy in place by 2028, highlighting the need for managed, high-performance architectures that balance innovation with compliance and control.