How to protect skills pipelines in the age of AI 

AI is scaling, but skills pipelines are under pressure, and digital leaders are feeling both realities at once. Technology teams are delivering faster with AI embedded across engineering, testing and delivery. Yet underneath that acceleration, new capability risks are emerging. 

As AI takes on more foundational work, the pathways technologists traditionally used to learn, practice, and progress are being reshaped. If workforce models are not redesigned alongside technology adoption, organisations risk creating a future where productivity rises but deep technical capability thins out. 

This tension sits at the heart of our recent Tech Flix film, ‘The AI Skills Paradox’: AI is Scaling, Skills are Not’, which explores the widening gap between the rapid rise of AI and the skills needed to harness it. It brings together perspectives from industry, education and government to examine how organisations can prepare their people for the future of technology. 

For a frontline operational view, we also spoke with Nash Squared CIO Ankur Anand, whose perspective reflects how these shifts are playing out inside technology teams today. 

The digital talent paradox inside tech teams 

AI adoption across technology functions has accelerated rapidly, but workforce readiness is moving at a different pace. 

The 2025 Nash Squared/Harvey Nash Digital Leadership Report shows that demand for AI capability continues to surge, with AI and machine learning remaining among the fastest-growing skills areas globally. Yet access to talent remains constrained, with digital leaders consistently reporting skills shortages in critical technology disciplines. 

This creates a structural paradox. 

Organisations are scaling AI delivery while facing persistent gaps in the very skills required to implement, govern and scale it effectively. 

Our recent Tech Flix film reinforces this divide, highlighting how AI is advancing faster than workforce preparedness, creating pressure on leaders to rethink how skills are built, not just how technology is deployed. 

The real risk to junior and mid-level roles 

Much of the external narrative focuses on AI threatening entry-level roles. Inside tech teams, the structural pressure (how the shape of a tech team is shifting) is more nuanced. 

Development copilots, automation tooling and internal knowledge systems now allow junior engineers to complete tasks that previously required several years of experience. 

As CIO of Nash Squared, Ankur explains, early-career productivity is rising sharply. Junior engineers can interpret requirements, generate code and build solutions far earlier in their careers than before. 

But the structural impact falls most heavily on the mid-experience layer. 

“It’s creating more risk for the people with mid-level experience compared to the more senior and experienced people as well as the juniors,” he notes. 

Work traditionally owned by engineers with two to five years of experience is being compressed. It is absorbed upward through AI-augmented senior oversight and downward through AI-enabled junior execution. 

This does not remove technology roles, but it does reshape career pathways. 

Without intervention, organisations risk narrowing the bridge between entry-level exposure and senior accountability. 

Productivity is rising & experience is evolving 

While risk exists, AI is also transforming what early-career technologists can achieve. 

Ankur points to initiatives delivered by engineers with less than a year of experience, including platforms launched within months. These were AI-enabled from inception and did not follow traditional development learning curves. 

“The impact of AI on fresh talent is very high. Their productivity is now almost as good as people with three to five years of experience.” 

Access to automation, coding copilots and internal data environments allows early-career engineers to contribute meaningful outputs faster than ever before. 

The challenge for digital leaders is ensuring that accelerated output still translates into deep expertise over time. 

How technical skills development is evolving 

AI’s impact is not uniform across the technology landscape. 

In modern product and platform environments, AI is embedded across delivery, accelerating coding, testing and documentation. Engineers are building AI-enabled solutions from day one, working in automation-rich ecosystems where delivery speed is significantly enhanced. 

As Ankur notes, when reflecting on recent initiatives, many programmes today are designed around AI from inception rather than layered in later. 

But the picture shifts in high-accountability, experience-led environments. 

In domains requiring deep expertise, risk ownership and judgment, human capability remains central. 

As Ankur explains, “Where you have high-skilled jobs with more experience and manual decision-making required, you can’t rely on junior or entry-level talent to take those judgment calls. But you may augment AI for experienced people so they can make faster decisions.” 

This creates two distinct capability pathways: 

  • AI-enabled engineering environments where automation drives productivity 
  • Experience-led environments where AI augments but does not replace human judgment 

Digital leaders must build strength across both. 

Governance is now a delivery priority 

As AI accelerates output, governance becomes inseparable from capability. 

“AI without governance is equally a big risk and can have unexpected consequences,” Ankur warns. 

AI-generated code can introduce vulnerabilities if deployed without architectural understanding. Security design, data privacy and resilience frameworks are not inherently embedded in AI outputs. 

Rapidly developed applications may function but fail under scrutiny or scale. 

For digital leaders, this reinforces four operational priorities: 

  • Security-first engineering principles 
  • Responsible AI training 
  • Human review layers 
  • Structured governance frameworks 

Speed must be balanced with safeguard design. 

Five leadership moves to protect the skills pipeline 

Protecting capability does not require slowing AI adoption. It requires designing workforce models that evolve with it. 

Drawing on insights from our recent Tech Flix film, the 2025 Digital Leadership Report and Ankur’s frontline perspective, five priorities stand out. 

1. Redesign early-career roles 

AI enables juniors to deliver faster, but learning must remain intentional. Exposure to architecture, testing and decision-making must sit alongside AI-enabled execution. 

2. Accelerate mid-level progression 

As delivery work redistributes, mid-career technologists must be supported to move into higher-value domains such as security, platforms and governance. 

3. Embed governance into engineering workflows 

Governance cannot sit outside delivery. Secure design, AI oversight and risk accountability must be built into day-to-day development. 

4. Build dual capability pathways 

Leaders must invest in both AI-enabled product skills and AI-augmented legacy expertise to sustain transformation. 

5. Design blended operating models 

Future tech teams will combine AI-enabled early talent, experienced engineers, governance capability and platform leadership rather than flattening structures entirely. 

The capability question facing digital leaders 

AI is already reshaping how technology teams operate. It’s accelerating junior contribution, redistributing mid-level work, and augmenting senior oversight. 

But long-term capability will not build itself. 

As Ankur emphasises, the priority now is workforce design. Leaders must ensure accelerated productivity today still produces the deep technical expertise organisations will depend on tomorrow.