Construction’s agentic AI transformation will be won by people, not technology

The UAE’s decision to transition 50 per cent of government sectors, services and operations to agentic AI within two years marks a decisive shift in the technology conversation. This is a move beyond digital tools that respond to instructions. Agentic systems can interpret an objective, plan a sequence of actions, coordinate tasks and execute with increasing autonomy. The government has matched that ambition with a programme to train 80,000 federal employees, recognising that a transformation of this scale cannot be delivered through technology alone. It requires people who understand how to direct, challenge and govern it.

For construction, the implications are significant. Every major project is shaped by thousands of connected decisions across planning, design, procurement, workforce deployment, safety, quality and handover. Agentic AI could help analyse schedules, anticipate delays, monitor supply chain risk, optimise resources and improve real-time visibility across complex sites. In an industry where a late decision can quickly become a programme delay or cost overrun, that capability has inherent value.

The greatest risk is not that construction companies will adopt AI too slowly. It is that they will adopt it without redesigning the work around it. An alert is only useful if a project team understands its significance and acts. A procurement recommendation must still be tested against supplier realities, contractual obligations and the wider project programme. A safety system must strengthen human accountability, not blur it. In a sector where decisions affect lives, assets and critical national infrastructure, judgement remains indispensable.

This is why the debate about whether AI will replace construction jobs is too simplistic. The more important question is how roles will evolve. Repetitive analysis, reporting and coordination will increasingly be supported by automation, while engineering judgement, problem-solving, stakeholder management and leadership will become more valuable. The World Economic Forum’s Future of Jobs Report 2025 highlights that the skills required across many roles will continue to evolve through 2030, while analytical thinking, resilience, leadership and collaboration remain among the capabilities employers value most. The future workforce will therefore need both stronger digital fluency and deeper human expertise.

Traditional job descriptions were built around relatively fixed responsibilities, whereas agentic AI requires employers to examine work at task level: what can be automated, what should be augmented and what must remain clearly human-led. Training must also move closer to the work. One-off courses will not keep pace with systems that are evolving rapidly. People need role-specific learning, access to live tools and opportunities to build confidence through real project challenges rather than abstract demonstrations.

Leadership and governance must also evolve with the technology. AI adoption cannot sit solely with innovation or IT teams. Engineers, planners, procurement specialists, safety leaders, legal teams, data specialists and HR must collectively define how systems are selected, tested and used. Employees also need the confidence to question outputs rather than defer to them. A culture that rewards blind adoption will create risk. A culture that encourages informed challenge will create value. The strongest organisations will not simply produce AI users; they will develop people capable of managing, supervising and improving AI-enabled work.

Workforce transformation

At Innovo, we see workforce transformation and digital transformation as the same agenda. Across more than 40 digital initiatives spanning design, BIM, procurement, safety, technical functions and site operations, the objective is not simply to introduce new tools. It is to build the skills, ownership and confidence required for those tools to improve project outcomes. That means involving functional experts early, embedding learning into live project environments and measuring adoption through operational value rather than activity alone. Technology only becomes transformative when the people closest to the work can shape how it is applied.

Construction will remain a fundamentally human-led industry. Its future will continue to depend on engineering expertise, site experience, collaboration and leadership. Agentic AI will not diminish those qualities; it will raise the premium placed on them. The organisations that lead this transition will not be those with the largest number of AI agents. They will be those that define the clearest relationship between machine capability and human judgement, redesign work around that relationship and invest in their people early enough to make it succeed.

The writer is the Chief People Officer at Innovo Group.

–  In association with Innovo Group