Only 34 per cent of organisations are genuinely reimagining their business with AI, according to Deloitte's 2026 survey of 3,000 senior leaders. For securities finance firms, the gap between boardroom AI strategy and production deployment is wide - and it's costing them speed, accuracy, and competitive ground.
Gartner's parallel findings reveal that 60 per cent of finance functions report some AI adoption, yet 91 per cent say the impact is low or moderate. The problem isn't the technology. It's the chasm between positioning and execution, and what's required to close it - clean data, redesigned workflows, and a culture that treats AI as an operating model, not a side project.
Ben Challice, CEO of Pirum, said the most important conversation with a client doesn't start with models. "It starts with a question: is your securities finance data standardised, complete, and flowing in real time across your operations?"
The data foundation gap
The use cases producing real returns in trading, post-trade, and collateral management follow a pattern. They're high-frequency, high-volume tasks - exception triage, break identification, collateral substitution matching - or prediction problems where historical data can anticipate costly outcomes. Settlement fail prediction in a T+1 environment is the clearest example: models trained on clean, real-time trade data flag likely fails hours before the settlement window closes, letting desks resolve issues without manual fire drills or penalty fees.
Citi's latest securities services survey of 537 firms found 86 per cent are already piloting generative AI in post-trade operations. BNY has 125 AI solutions in production through its Eliza platform, including AI agents that autonomously validate payment instructions. State Street's Alpha platform delivered a 25-times productivity improvement in post-trade data exception processing - not because of a more sophisticated model, but because the system could distinguish the 250 genuine problems from 31,000 flagged items. These results rest on infrastructure that already had lifecycle connectivity and real-time data, something only 43 per cent of firms are fully confident they possess, per a Riverbed survey of 1,200 financial services decision-makers.
Closing the gap between pilot and production
Speed of deployment now matters more than it did in any previous technology cycle. Agentic AI - assistants that reason through multi-step problems and act without human initiation - is reaching post-trade workflows. Wells Fargo has AI agents handling complex foreign exchange post-trade inquiries across its corporate and investment bank. IDC research from March 2026 shows 80 per cent of capital markets firms now cite building AI agents as their top IT spending priority, and early adopters report 2.3-times ROI within 13 months.
"At Pirum, we are not approaching AI as an experiment," Challice said. "It is a fundamental and complementary shift that raises the bar for how we build, support, and operate." The firm has embedded agentic AI engineering across product development cycles and redesigned onboarding and client support around AI-augmented workflows. For traders and operations teams, the insight - a predicted deadline conflict, a counterparty re-rate signal, an auto-generated recall instruction - must surface inside the tools they already use, not a separate dashboard.
Client interactions that anticipate, not just report
The same standard applies to how a post-trade platform interacts with its clients. HSBC AI Markets, which gives institutional clients real-time access to research and pricing via natural language processing, points to a model: data turned into something actionable at the point of decision. A reconciliation exception report delivered after the fact informs a client. A flag delivered 12 hours before settlement, with a probable cause and pre-populated resolution pathway, serves one. The constraint is rarely the model - it's the timeliness and cleanliness of the underlying data.
When support teams arrive at each interaction already aware of a client's operational patterns and recent pressure points, cases resolve faster and with less effort. Outreach becomes proactive because data generates intelligent insight on demand. This shift from reactive reporting to anticipation is where client relationships gain depth, and it is a direct result of foundational AI for Finance strategy that extends well beyond a chatbot interface.
Culture: the hardest part of becoming AI-first
Citi has deployed AI tools to 140,000 employees and made AI fluency training mandatory for 175,000. Goldman Sachs is deploying thousands of autonomous AI coding agents, with its CTO projecting three to four times productivity gains. BNY's AI agents have defined identities and access controls, moving humans from executing tasks to training and nurturing those agents. These are cultural shifts, not just technical ones.
BCG has found that 70 per cent of AI's impact comes from changing how work gets done, not the technology itself. McKinsey's analysis of banks generating the most AI value identified cross-functional teams with clear ownership as the differentiator - not model sophistication, but whether the people closest to the problem are empowered to act. For AI for Executives & Strategy, this means treating AI capability as a shared responsibility, not a function that lives in a specialist unit.
"Becoming AI-first in securities finance is not a project with a go-live date," Challice said. "It is a continuous operating model evolution, sustained by a culture that treats AI capability as a shared responsibility rather than a specialist function."
Why this matters for finance professionals
Firms that wait for perfect data or a fully formed AI strategy before moving will fall behind those that start with the cleanest, most real-time data they have and redesign workflows around agentic AI now. The firms already seeing 25-times productivity gains and 2.3-times ROI didn't win with better models - they won by connecting AI to the operational plumbing and creating an organisation where AI is everyone's job. For trading desks, post-trade teams, and collateral managers, the practical next step is to audit the completeness, standardisation, and speed of your data flows. That audit determines what AI can actually do for you.
Your membership also unlocks:
