A potential pitfall with agentic AI? Settling for the easy wins

Irish firms are embracing agentic AI, but real transformation demands more than early efficiency wins, writes Pratuesh Raj

AI robot surrounded by charts

Irish businesses are embracing agentic artificial intelligence (AI), but few are unlocking its full transformative potential.

AI agents are reshaping the way work gets done and, in Ireland, some organisations adopting agentic AI are seeing tangible benefits.

The top reported gains include increased productivity (53%), cost savings (38%) and faster decision-making (31%).

Other areas of impact—such as innovation in products and services (20%), improved profitability (16%), and faster time to market (16%)—indicate the technology is starting to move beyond delivering efficiency gains toward strategic value.

But here’s the potential pitfall: settling for too little. Early wins, such as productivity boosts, are important, but they stop short of transformation.

Many employees are using agentic features built into enterprise apps to speed up routine tasks such as updating records and answering questions. That’s helpful, but it’s not game-changing.

To unlock a bigger impact, companies need to build multi-agent models—systems of AI agents that work together to deliver results across complex, cross-functional workflows and vendors.

Getting there will require addressing the biggest barrier: mindsets.

Right now, people, including senior leaders, are holding AI agents back. When we look at the top challenges cited by Irish executives, concerns like cybersecurity and cost often dominate. But these are solvable.

AI agents can be made secure, and a well-designed implementation can pay for itself within months.

Here’s how Irish companies can go further with agentic AI:

  1. Focus on people. AI agents deliver the most value when they help workers rethink — and even reimagine — their day-to-day work. Equip teams with the skills to integrate agentic AI into their routines and encourage them to use the technology to create more value.
  2. Orchestrate and integrate. Using a few AI agents in isolation on existing applications won’t move the needle. Leading organisations adopt the tool across a range of processes, using AI enterprise command centres to connect agents, coordinate workflows, and accelerate rollout.
  3. Prioritise trust. As AI agents advance, they’ll make more decisions autonomously. To build stakeholder confidence, anchor your AI agent strategy in a responsible AI foundation.

Pratuesh Raj is Director of Technology Consulting at PwC Ireland