Three things that leaders discover when they truly begin to interact with AI
Dr Jeremy Bradley, Chief AI Officer of Cambridge Spark, shares three key lessons that financial sector leaders discovered when seriously tackling AI adoption in their organisations.
"No plan survives first contact with the enemy." Helmuth von Moltke the Elder.
There is a version of AI adoption that looks impressive, planned in a slide deck, but quickly falls apart in execution. Models that no one trusts. Outputs that no one can explain or justify. Productivity gains in one corner of the business that quietly mess something up three steps down the line. It has become obvious that the difference between AI that delivers impact and AI that doesn't comes down to three things: rigour, trust, and a genuine, open assessment of risk.
Rigour. The way in which AI processes and tools are tested and validated to perform under critical conditions.
Trust. Trust must be earned from the stakeholders who will be accountable for the tool in service.
Risk. The appreciation that AI tools do not deliver deterministic outputs, meaning there is a quantifiable risk that an output will be incorrect or sub-optimal at some point.
Last week I had the privilege of helping to deliver three AI leadership courses for executives in the financial services industry. These were not first-time awareness sessions; these were senior leaders on our Leading with AI programme, seriously grappling with how AI fits into their organisations. What struck me was not the engagement and enthusiasm — though there was plenty of both — but the nature of the questions being asked. Across all three courses, the same themes emerged.
Here is what those leaders were thinking about AI, and what it means for getting transformation right.
1. "Will AI make my team forget how to think?"
Underlying the excitement over AI's productivity potential is an anxiety: what happens to human capability when AI begins to make significant technical contributions? It is an observed phenomenon in software engineering: junior engineers who over-use code generation see their personal capability degrade.
So it is a legitimate question. If professionals stop executing critical tasks themselves (data processing, complex analysis, code debugging), does that skill quietly atrophy? And if it does, what is the fallback when the system fails?
The answer is not to slow down AI adoption, but to be smart about it. For some tasks, total automation is the right approach. For others, where AI is still needed for all the benefits it brings — think surgical procedures, landing an aircraft, or any mission-critical decision — maintaining human capability as a genuine fallback is essential, it is necessary governance. The conversation every leadership team needs to have is not "should we use AI here?" but "if the AI fails here, can we still land the plane?"
2. "Our data is a mess: AI will have to wait."
Many businesses believe their technical debt is a showstopper. Years of legacy architecture, siloed platforms, and inconsistent data standards—surely that needs sorting before AI can deliver value?
Not necessarily. In fact, AI can actively accelerate navigating complex data environments and restore confidence in data and what it tells you. Proof-of-concept models can be quickly deployed using ad hoc interfaces and data-cleaning agents, meaning you do not need a pristine data estate to start delivering impact. The goal is not perfection before you start; it is to learn quickly, build trust in your data sources as you go, and understand the quality and provenance of what is feeding your models.
Technical debt is certainly real. But it is now a reason to use smarter approaches, not a roadblock to progress.
3. "We optimised one team and broke another."
This one drives a systems-thinking approach. Leaders arrive thinking about AI within business functions. They leave thinking about AI across the entire business.
Making one function hyper-productive without enabling downstream teams to absorb that output does not create value—it creates bottlenecks, frustration, and risk. A sales team generating ten times more leads is a liability if operations, legal, and compliance cannot scale their execution to match.
Genuine AI transformation requires a holistic view of your business flows. This means collaborative governance, cross-functional engagement, and a clear assessment of the end-to-end process before you optimise any single part. Done right, the ROI transforms the organisation. Done in silos, you simply push the problem downstream.
Share this post:
