Redesigning jobs and skills for the AI era

AI adoption is advancing. Organisations are introducing new tools and technologies, automating tasks, and investing in training. Yet their approach to work, including the people systems and processes that support it, often remains unchanged.

As work evolves, skills frameworks, career paths, and performance management processes are becoming outdated. These structures were designed on the assumption that people perform every task. That assumption no longer holds, requiring Irish employers to rethink how work is designed and delivered, how skills are developed, and how this change is managed.

AI is changing the work, not just the tools

AI is moving beyond experimentation and becoming part of everyday work. For many organisations, the focus has shifted from whether AI can improve productivity to where it fits within the operating model, which tasks it should change, and what those changes mean for people. In most cases, AI does not replace an entire role at once. It changes the individual tasks within it.

A role may include tasks that can be automated, tasks that can be improved through collaboration between people and AI, and tasks that should remain human-led because they depend on judgement, trust, context, relationships or accountability. Leaders therefore need a more detailed understanding of work than many existing job descriptions, competency frameworks, and career paths provide.

The CHRO’s architectural blueprint: Redesigning jobs and skills for the AI era argues that most organisations still rely on job architecture designed for a pre-AI world. Roles were largely defined by the tasks people performed rather than the outcomes they were expected to deliver. This disconnect can leave employees uncertain about how their roles are changing, make it harder for managers to set expectations, and lead talent systems to reward work that is becoming less important to value creation.

And work is changing faster than roles

Many organisations have invested heavily in AI over the past two years, from pilots and use cases to training programmes and productivity tools. Yet the way work is organised has often remained largely unchanged. Roles are evolving, but how they are defined isn't keeping pace.

Consider a recruiter. AI can help source candidates and draft communications, giving the recruiter more time to build relationships, assess suitability, and support hiring managers in making better decisions.

A financial analyst can use AI to produce initial drafts of reports in minutes rather than hours. This creates more time to interpret results, challenge assumptions, communicate insights, and influence business decisions.

A customer service adviser may work alongside an AI assistant that resolves routine queries before they reach the adviser. This can improve response times while allowing the adviser to focus on more complex situations that require empathy, judgement, and problem-solving.

New responsibilities are also emerging informally. An ‘AI champion’, ‘super-user’ or ‘go-to AI person’ may spend significant time helping colleagues adopt new tools, testing use cases, and sharing good practice. This work is often performed alongside the person’s existing role, with little clarity about expectations, recognition, or development opportunities. Over time, this creates a disconnect between the work people perform and how organisations hire, develop, reward, and support them. The effects may include:

• uncertainty about responsibilities;

• inconsistent use of AI across teams;

• varying standards of quality and oversight; and

• uncertainty about how careers will develop as roles evolve.

This disconnect also makes workforce planning more difficult. Viewing jobs as fixed can obscure where AI is reducing effort, where demand for human expertise is increasing, and where new capabilities are emerging. Understanding AI’s impact requires organisations to look beyond job titles and examine the work itself.

Start with the work that creates value

The answer isn't to rewrite every job description overnight. A more practical approach is to focus first on the roles and activities that contribute most to organisational performance, customer or service-user outcomes, and risk management.

For each of these roles, ask:

• What outcomes is the role responsible for?

• Which activities are being reduced, accelerated, or added?

• What capabilities will people need to succeed?

• How should performance be measured when AI forms part of the process?

• What safeguards are needed to maintain quality, accountability and trust?

Organisations can then examine how that work is performed and where AI could change it. This isn't about adding AI to every process; it's about redesigning work to create greater value while keeping the focus on outcomes.

Without this clarity, AI adoption may become fragmented, with teams working in different ways and applying inconsistent standards across the organisation.

Redesign career paths as work changes

As AI becomes part of everyday work, organisations will need to rethink how people develop expertise.

With AI support, less experienced employees may be able to take on more complex work earlier in their careers, challenging traditional assumptions about progression. At the same time, as some routine activities disappear, organisations must find new ways for employees to build the experience and judgement those activities once provided.

This raises several questions:

• How will organisations develop the experts they need in the future?

• How will they identify potential?

• What experiences will replace traditional learning pathways?

• How will employees build credibility and confidence as their roles evolve?

These questions are particularly relevant in Ireland, where organisations are competing for specialist digital talent while also seeking to reskill and redeploy employees as workforce needs change.

AI is accelerating skills change

In Ireland, we're starting to seeing a shift in how organisations think about the future of skills. PwC’s 2026 Global AI Jobs Barometer shows that hiring of AI skills in the Irish market doubled between 2024 and 2025.

Globally, the skills required for the jobs most exposed to AI are changing more than twice as quickly as those required for the least exposed jobs and this disruption is accelerating. The difference in the rate of skills change between the most and least AI-exposed roles increased from 25% over the period from 2019 to 2023 to 116% from 2019 to 2025.

The research also found that productivity growth was 40% higher in the companies most exposed to AI than in those least exposed. These companies also recorded faster growth in headcount and wages. AI exposure is increasingly associated with workforce reinvention and growth, rather than efficiency alone.

Build the skills to work effectively with AI

The 2026 Global AI Jobs Barometer has highlighted how quickly skills are changing and, for organisations in Ireland, the response cannot be limited to just technical training.

Employees need the confidence to use AI tools, but they also need the skills to apply them effectively. These include:

• exercising judgement when AI-generated outputs are incomplete or incorrect;

• asking the right questions and defining problems clearly;

• communicating recommendations and decisions;

• collaborating across disciplines;

• understanding risk, ethics and, accountability; and

• recognising when human intervention is needed.

For many employees, the most valuable skill will not be the ability to build AI systems, but the ability to work effectively alongside them.

Managers face a different challenge. They need to redesign work, set clear expectations, support employees through change, and define what good performance looks like when AI forms part of the process.

General awareness programmes can help people understand AI, but understanding alone doesn't necessarily change how they work. Training is more likely to be useful when it develops practical, role-specific skills that employees can apply to their day-to-day responsibilities.

AI can enable people to focus on higher-value work, improve productivity, and support new sources of growth. However, adding it to existing ways of working will not deliver those benefits on its own. Leaders must determine where work should change, equip people with the necessary skills, and ensure that AI improves outcomes rather than simply accelerating existing processes.

Next steps

1. Understand where work is changing

Start with a small number of business-critical roles or processes and examine the work at an activity level. Identify which activities can be automated, which are best performed by people working with AI, and which should remain human-led. Focus on how the work is changing in practice, rather than relying on assumptions based on job titles or organisational structures.

2. Redesign roles around outcomes

Review role profiles, accountabilities and decision rights in terms of the outcomes each role must deliver. As AI takes on some activities and changes others, clarify where human judgement, oversight and accountability remain. Avoid adding AI-related responsibilities to existing roles without considering how the role itself, and the value it creates, should change.

3. Build skills for AI-enabled work

Look beyond technical training to identify the skills people need to work effectively with AI. These may include critical thinking, judgement, problem definition, evaluation of AI-generated outputs, communication, ethical decision-making, and responsible use. Incorporate these skills into learning, development, and career pathways so that they become part of how the organisation develops its people.

4. Align performance, progression and reward with new sources of value

As AI changes the relationship between effort and impact, review how performance is measured and recognised. This may require greater emphasis on outcomes, quality, innovation, collaboration, judgement, and responsible use of AI. Career paths may also need to change as traditional routes to expertise evolve. Progression should reflect the skills and contribution that matter in an AI-enabled organisation.

5. Treat work redesign as organisational change

Help employees understand how their work is changing, why it is changing, and what it means for them. Be open about which activities will be automated, supported by AI, or remain human-led, and give people opportunities to develop the skills and confidence they need. Clear leadership and regular dialogue can help maintain trust and consistency as new ways of working take shape.

Turning AI ambition into workforce change

Organisations now need to translate their AI ambitions into practical changes to work, roles, and skills. PwC can help you assess how roles are evolving, redesign job and skills architecture, align talent processes, and support people to use AI confidently and responsibly. To explore what these changes could mean for your organisation, contact a member of our team.