From GenAI Enablement to Enterprise Enablement: The CTO Function Reconsidered 

AI may be the catalyst, but the larger opportunity is a hands-on capability that helps capital markets technology organisations learn, adapt and improve at scale.

Large financial institutions do not lack capable technologists, strong engineering leads or informed business experts. Most already have deep expertise across development, architecture, cloud, data, quant, support and control functions. The problem is not that good ideas are missing. It is that insight is often trapped in local contexts, applied within one team or programme, and not translated into wider value across the organisation. 

At the same time, CIOs and CTOs are expected to guide their technology organisations through constant change: modern engineering practices, cloud transformation, growing technical debt, rising cost pressure, tighter control expectations, and now GenAI-enabled ways of working. That is difficult to do from governance forums, architecture principles and reporting lines alone. 

The CTO function needs to get closer to the work. 

Not to take control away from system owners or team leads. Not to centralise every technical decision. But to create the visibility, support and feedback loops that allow distributed expertise to drive better outcomes across the firm. 

GenAI Has Made the Gap Harder to Ignore 

GenAI is not the whole story, but it has made the underlying organisational problem much more visible. 

Across financial institutions, AI tooling is already changing how analysis, development, support and documentation are performed. Teams are using AI to understand legacy systems faster, draft specifications and design notes, generate tests, improve support diagnostics, explain complex logic and reduce dependency on tribal knowledge. But adoption is uneven. It depends on approved tooling, local confidence, control expectations and the realities of each delivery environment. 

Many of our clients are already responding by creating AI enablement teams, champion networks or centres of excellence. That is a sensible first step, and many of those groups are already adding value. 

But they are often narrow by design: focused on AI tooling, policy interpretation, awareness, or a limited set of use cases. In doing so, they risk missing the larger opportunity. 

The challenge exposed by GenAI is not just how to adopt AI tools safely. It is how to help a complex technology organisation absorb new capabilities, share learning, close skills gaps and translate local innovation into wider enterprise value. AI enablement is part of that answer, but it should not be the whole model. 

The Real CTO Challenge Is Not More Control 

In many firms, the modern CTO role risks becoming too distant from real delivery. 

It can become associated with standards, target-state architecture, review boards and oversight of change, while the practical realities of adoption sit elsewhere: in delivery teams, platform teams, engineering leads, quants, support analysts and business SMEs. 

But in a complex technology organisation, the most useful information is often local. It sits with the people closest to the systems, constraints, inefficiencies and opportunities. No central function can or should try to replace that distributed knowledge. 

The role of the CTO should therefore be less about pulling decisions to the centre and more about making distributed expertise work at enterprise scale. 

That means giving CIOs and CTOs better ways to understand what is really happening across the organisation: where useful practices are emerging, where technical debt is building, where tooling gaps are slowing teams down, where controls are being interpreted inconsistently, and where ideas in one area could materially benefit another. Just as importantly, it means giving teams confidence that someone is listening, that useful ideas will travel, and that support is available when they need to adopt new tools, improve ways of working, or solve problems they do not have the spare capacity to tackle alone. 

What the CTO Function Should Actually Look Like

A credible CTO function cannot rely on standards and reporting alone. It needs hands-on capacity to understand how teams really work, where change is being blocked, which capabilities are missing, and which improvements are worth scaling. 

In practice, that means a small, mobile, technically credible enablement capability sitting within the broader CTO or CIO organisation. This team would work directly with existing technology teams to understand what they are doing and trying to achieve, what is working and what is not, which tools, practices and workflows they rely on today, what they wish they could use, what they are not aware of elsewhere in the organisation, where risks and inefficiencies are quietly accumulating, and where a small intervention could unlock disproportionate value. 

The purpose is not inspection for its own sake. It is to acquire real information and turn it into practical action. 

That practical element is critical. Most teams do not resist change because they are unwilling. They resist it because they are busy, accountable for live systems, and already balancing delivery commitments, support obligations, architectural constraints and control requirements. Even when a team knows it should improve its tooling, automate a manual process, adopt a better engineering pattern or modernise part of a platform, the cost of change is real. The work still has to be evaluated, implemented, embedded, documented and taught. Existing teams often cannot absorb all of that on their own. 

This is where the CTO function needs hands-on resource. It should provide additional advisory and delivery capability to help system owners and team leads do things in their area of responsibility that they might not otherwise have the bandwidth, context or support to pursue. The role is not to replace local ownership, but to make change more achievable: helping teams try new tools, techniques and processes in a safe and practical way, embedding them into the team’s way of working, and then stepping back once the capability is established. 

The existing system owner, engineering lead or team manager should remain accountable for the system and its outcomes. The CTO enablement function does not take responsibility away from them. It strengthens their ability to exercise it well. 

For a team lead, that should feel valuable. Leading a system or team often means losing some of the direct access to a broader peer group that individual contributors still have within the team. It becomes harder to stay close to emerging tools, compare approaches with other areas, or carve out time to explore alternatives while still meeting delivery commitments. A good CTO enablement capability helps close that gap by bringing additional senior thinking, extra delivery capacity, broader organisational context and stronger routes into peer learning.

What the CTO Group Would Do in Practice 

A hands-on CTO enablement team would not be another innovation lab detached from delivery, nor a central review board operating at a distance. 

It would listen, interpret, challenge, connect and help implement. 

In practice, that means working with teams to understand systems, pain points, goals and constraints; identifying useful tools, practices or process changes that could help; assessing what is feasible within the organisation’s policies and operating model; providing additional capacity to trial and embed improvements; filling capability gaps where specialist support is temporarily needed; upskilling local teams through joint execution rather than classroom training alone; capturing useful patterns and lessons for wider benefit; and surfacing common issues, opportunities and risks to senior technology leadership. 

This is much closer to a delivery capability than a traditional staff function. 

What This Looks Like for Engineering Teams 

For architects, engineering leads, support teams, quants and business-aligned technologists, this model should feel supportive rather than intrusive. 

It should help them maintain their edge. It should bring access to scarce expertise when needed. It should create a route for ideas and frustrations from the ground to be heard, understood and acted on. And it should help teams discover what else is happening in the organisation that they could reuse, learn from or align to. 

That is increasingly important now that delivery depends on a broader set of capabilities at once: engineering, architecture, data, controls, cost, resilience, documentation, observability and evolving AI-enabled workflows. Teams do not just need permission to improve. They need practical support to improve well. 

What This Looks Like for CTOs and CIOs 

For senior technology leadership, the value is different but equally important. 

A hands-on enablement function provides a mechanism for understanding the estate at greater depth. It surfaces what is actually happening, not just what is being reported. It reveals where strong practices deserve scaling. It identifies debt, duplication, tooling gaps and control friction earlier. And it gives leadership practical levers for influence beyond policy and organisational structure alone. 

This is especially important where leadership wants to empower teams rather than centralise every decision. Empowerment only works well at scale when local autonomy sits within a framework of transparency, alignment and feedback. The aim is not to make every technical decision centrally. It is to make good local decisions more visible, more connected and more consistently aligned to wider business and technology goals. 

Why This Matters Beyond AI 

GenAI may be the current catalyst, but the need is broader. 

The bigger opportunity is not to build a narrow AI enablement function in isolation, but to use GenAI as the trigger for a broader enterprise enablement capability. 

The same organisational challenge appears in cloud, DevOps, support modernisation, observability, platform engineering, cost governance, model tooling and data platform change. In each case, firms need a way to absorb new capabilities without fragmenting delivery, increasing hidden risk or creating fresh inefficiencies. 

This should not be seen as an AI function. It is an enterprise enablement capability for technology change more broadly. AI simply makes the need more urgent because the rate of change is faster, the range of use cases is wider, and the gap between what is possible locally and what is adoptable safely at scale has become harder to ignore. 

A Better Way to Think About the CTO Function 

The most useful way to think about the CTO function now may be this: not as the place where all important technical decisions are made, but as the enterprise mechanism that helps the organisation understand itself, learn faster, and act more coherently. 

In that model, the CTO is still “chief”; but not because everything flows upward for approval. Rather, because the function helps turn distributed expertise into enterprise progress. 

Conclusion 

The future CTO function is less about centralising authority and more about enabling intelligent, aligned action at scale. 

Financial institutions already have strong experts. What they often lack is a practical mechanism for hearing what those experts are seeing, amplifying what is working, responding earlier to risk and debt, and helping teams evolve without losing control. 

That requires more than governance. It requires hands-on capacity to sense, support and shape change across the organisation. 

GenAI has made that need more urgent. But it did not create it. 

At Phi, we recognise the same challenge in our own delivery model. We are putting structured programmes in place to strengthen our consultants’ AI-enabled delivery skills, with role-based learning, controlled sandboxes, engagement-specific training and practical knowledge sharing, so that teams can apply GenAI responsibly and effectively in client environments. Our approach is explicitly delivery-led: aligned to client-approved tools, policies and operating constraints, and focused on improving quality, speed and resilience in live work rather than treating AI as a separate practice. 

The firms that succeed will be those that combine empowered local expertise with stronger enterprise visibility, reusable learning and practical enablement. That is how CIOs and CTOs gain the information and levers they need to influence their organisations at depth. And it is how teams get the support they need to improve the systems the business depends on.