Ben Hunter, Senior Director of Financial Services, Gigamon
AI can deliver a decisive edge in the highly competitive financial services industry. Retail banks and other financial institutions (FSI) are looking to reap the transformative benefits of AI to streamline processes and operations, improve customer service and tap into predictive analytics.
AI adoption is rapidly accelerating. The Bank of England reports that, in just two years, AI uptake increased by 34%, with 75% of FSIs now using AI, and a further 10% planning to use it over the next three years.
The sense of urgency is driving rapid progress, but it is also creating explosive data growth, increasing complexity, and introducing new risks in the hybrid cloud infrastructures that underpin AI deployments.
Monitoring the data being fed to AI models across the network is becoming more difficult. Shadow AI is expanding behind the scenes, model behaviour is often unclear, and data is moving in ways that traditional tools can’t fully capture. Security and IT teams are realising that what was once a static and controlled environment is now dynamic and fragmented, making data visibility hard to come by and risks harder to define and contain.
The hidden risks of AI adoption
The security challenges tied to AI adoption are growing more defined with each deployment. One of the most urgent challenges that financial institutions face is shadow AI: the unsanctioned use of AI tools and deployments that operate outside the view of security and IT teams. Though these instances are rarely malicious, unauthorised use of AI tools still poses a threat to FSIs’ security and compliance, leading to potential data breaches and fines. Without monitoring AI activity, it becomes nearly impossible to track what data is accessed, how it’s used to train models, and where it ultimately flows.
In tandem, AI is putting pressure on the network itself. Generative AI (GenAI) models and automated workflows are producing an unprecedented increase in internal traffic. According to the Gigamon 2025 Hybrid Cloud Security Survey, one in three organisations have seen traffic volumes double over the past two years due to AI workloads, and most security leaders admit they’re making trade-offs to keep pace.
The increase in traffic creates blind spots in the hybrid cloud infrastructure that traditional monitoring tools struggle to keep up with. The same survey reveals that nearly half (46%) of Security and IT leaders lack clean, high-quality data needed to secure AI traffic. Without clear guardrails in place, GenAI systems can introduce risks, undermining data integrity and trust. Complete visibility across the hybrid cloud infrastructure, where critical AI models are located, is essential to reap the benefits of AI safely.
Visibility is the key to regaining control
As AI reshapes FSIs’ infrastructure, the need for complete visibility of data is becoming a foundational security requirement. Security teams must be able to see which workloads interact, how data flows across environments, and where risks may emerge. This is critical in an age where AI systems introduce more automation, generate new traffic patterns, and rely on interconnected devices.
Without complete visibility, even well-governed deployments can carry hidden risks. A fragmented view of the network makes it harder to respond to a wide range of potential issues, from misconfigurations to data moving via unexpected channels, and threat activities blending into everyday operations.
In this new landscape, innovative security strategies are emerging to achieve real-time visibility into GenAI activity and address shadow AI usage, enforce policy, and optimise ROI in AI deployments.
Deep observability provides actionable insights to reduce risk and improve governance by integrating network-derived telemetry, including packets, flows, and metadata, with existing metrics, events, logs, and traces (MELT) data. Having this complete visibility of all data in motion enables Security and IT teams to detect and monitor threats that traditional tools might miss. Security leaders agree, with 88% identifying deep observability as critical to securing AI deployments, according the 2025 Hybrid Cloud Security Survey.
Adopting more integrated visibility strategies allows FSIs to gain a clear, real-time picture of their hybrid cloud environment. Merging infrastructure, application, and network data enables security teams to detect issues earlier, respond faster, and build trust in how their AI systems are managed.
The risks associated with AI evolve faster when left unmanaged. Security leaders need to take a proactive approach, leveraging deep observability to detect and monitor GenAI and LLM engines to stay in control.
