Banks are accelerating investment in AI-driven surveillance, but many are struggling to turn that ambition into operational reality, according to new research from 1LoD, commissioned by Shield, a global communications risk management company purpose-built for financial institutions.
Despite increased investment and accelerating AI adoption, the industry’s biggest surveillance challenges have moved surprisingly little since 2024. More than 90% of institutions continue to operate in alert-heavy environments, and false positives remain the industry’s top-rated surveillance challenge, with no meaningful improvement since 2024. The findings point to a widening execution gap: Many institutions remain constrained by legacy systems, fragmented data and operating models designed for a different era, limiting their ability to translate AI investment into transformation at scale.
The findings are drawn from “The Future of Surveillance: Turning AI Ambition into Operational Reality,” by 1LoD, a specialized conference and intelligence provider focused on non-financial risk and compliance in banking, based on responses from senior surveillance and compliance professionals across financial institutions. The report compares its findings against 1LoD’s 2024 benchmarking survey to track how the industry’s challenges have evolved.
False positives remain the top-rated challenge, cited as highly significant by 52% of firms, followed by budget and resourcing constraints (41%), limited access to quality data (37%) and legacy or outdated systems (37%). By contrast, only 7% of respondents rated regulatory pressure as highly significant, underscoring one of the report’s central findings: For most institutions, the primary barrier to better surveillance is no longer regulatory uncertainty, but operational execution.
Shield’s own deployment data, cited in the report and drawn from evaluations spanning millions of communications, shows roughly a threefold reduction in alert noise compared to legacy systems. In direct comparisons, this has corresponded with up to 44% higher accuracy and about three times more actionable escalations, indicating that fewer, better-targeted alerts can lead to more effective risk identification.
“AI ambition is everywhere right now. What’s missing is execution,” said Shiran Weitzman, Co-Founder and CEO of Shield. “Investment is accelerating, but the fundamentals still matter. AI cannot deliver its full potential when surveillance is constrained by fragmented data, legacy infrastructure, and alert-heavy operating models. Closing that gap is what will separate firms that simply deploy AI from those that fundamentally improve how risk is detected and understood.”
The report also identifies what separates leading compliance organizations from those still struggling to operationalize AI and translate that investment into measurable improvements in surveillance and greater regulatory readiness. These firms are moving away from static, rules-based detection toward adaptive models that learn from behavioral patterns, unifying fragmented data and workflows into integrated platforms, and tying surveillance investment to measurable reductions in manual review. The report frames this operational shift as the deciding factor in whether AI investment improves detection outcomes.
The report concludes that surveillance is moving toward a model defined by integrated data, context-driven detection, reduced operational noise and measurable outcomes, and that institutions that close the gap between available technology and actual practice stand to lower costs while improving true risk detection.
The full report, “The Future of Surveillance: Turning AI Ambition into Operational Reality,” is available here.
Sign in to get access to this free resource, and all of our whitepapers and reports.
Download this content today!
Register Now Already registered? Click here to Log In
