From predictive algorithms to autonomous decision-making, Agentic AI is transforming how organisations in the Middle East and Africa design, deliver, and measure customer experience — but only if leaders can balance innovation with trust, talent, and ROI.
The New Lexicon of Enterprise Technology
Just a few years ago, AI was a buzzword cautiously dropped into conference conversations and strategy papers. Today, it’s as commonplace as “cloud” or “data analytics” in boardrooms, budgets, and business plans. Yet, as Basil Ayass, Interim Leader for Africa at Google Cloud, reminded the audience during the recent Enterprise IT World MEA panel, we are already moving into a more sophisticated, more autonomous, and potentially more disruptive phase: Agentic AI.
“AI entered our vocabulary relatively recently,” Ayass said, “but Agentic AI — AI that can act autonomously towards goals — is now entering our operational reality. It’s the difference between a GPS telling you the route and a self-driving car actually taking you there.”
Across the Middle East and Africa (MEA), the speed of AI adoption has been breathtaking. Government-backed digitalisation programs, pandemic-driven acceleration, and the rise of AI-powered customer interactions have created a market ripe for innovation. But the next frontier is not just about adopting AI — it’s about rethinking how machines make decisions, and how that reshapes customer experience (CX) at every touchpoint.
“AI has entered our lexicon in just a few years — but Agentic AI is about to enter our boardrooms.”
– Basil Ayass, Interim Leader for Africa, Google Cloud
From Novelty to Necessity
When machine learning (ML) first entered mainstream business use, it largely supported human-led decision-making. Predictive models could suggest marketing campaigns, identify fraud patterns, or forecast supply needs. But they still needed human oversight for execution.
Agentic AI changes that equation. “With Agentic AI, you’re not just analysing — you’re authorising,” said Ebrahim Kamalzadeh, CIO, Al Nabooda. “These systems can take actions in real time without waiting for human approval, whether it’s resolving a customer issue, rerouting a delivery, or offering a personalised discount.”
For businesses in competitive sectors like banking, telecom, and retail, this capability is a game-changer. Instead of reacting to customer needs, companies can anticipate and address them before the customer even realises the need exists. A credit card fraud detection system, for example, could not only flag suspicious activity but also automatically freeze the account, issue a temporary virtual card, and alert the customer — all within seconds.
“Customer experience isn’t just about satisfaction; it’s about building enduring trust in a digital-first world.”
– Ebrahim Kamalzadeh, CIO, Al Nabooda
What Is Agentic AI?
Agentic AI refers to systems that operate with a degree of autonomy, making decisions and taking actions aligned with pre-set objectives. Unlike narrow AI, which performs a specific task, or general AI, which aims to mimic human intelligence broadly, Agentic AI is goal-oriented and task-executive.
Basil Ayass described it succinctly: “Think of it as AI with the power to act, not just advise.”
In CX terms, this means an AI system could:
- Handle end-to-end customer support queries without human intervention
- Proactively offer solutions before problems escalate
- Adapt its actions based on the customer’s historical behaviour and real-time context
In MEA, where cultural and linguistic diversity adds complexity to customer engagement, Agentic AI can also navigate multilingual interactions, cultural nuances, and regulatory requirements at scale.
“You can’t deploy autonomous systems without ethical guardrails.”
– Sandeep Mahindra, Senior Enterprise Solution Architect, GBM
Customer Experience in the Age of Autonomy
For years, CX teams have focused on omnichannel consistency — ensuring that customers receive the same quality of service whether they interact via mobile app, call centre, website, or in person. Agentic AI raises the bar.
“Customer experience is no longer about ‘meeting expectations’,” said Sandeep Mahindra, Senior Enterprise Solution Architect, GBM. “It’s about creating moments of unexpected delight, where the system anticipates your needs and resolves them before you even have to ask.”
In practical terms, that could mean a travel booking platform detecting that a traveller’s flight is delayed and automatically rebooking the next connection, notifying the passenger, and adjusting hotel check-in — all before they land.
The result? Reduced friction, increased loyalty, and higher customer lifetime value.
“The winners in the AI era will be those who combine speed with sensitivity.”
– Fahad Qureshi, Director -IT, Talabat
Balancing Innovation with Trust, Ethics, and Control
The autonomy of Agentic AI is both its strength and its risk. In markets with strong data privacy laws, such as the UAE’s Personal Data Protection Law (PDPL) or South Africa’s Protection of Personal Information Act (POPIA), enterprises must ensure AI actions comply with legal boundaries.
“Autonomy without accountability is a recipe for disaster,” warned Fahad Qureshi, Director -IT, Talabat. “Just because the system can act doesn’t mean it always should.”
Building trust in AI-driven CX means:
- Establishing clear ethical guidelines for AI decision-making
- Creating human oversight frameworks for high-impact actions
- Maintaining transparency with customers about when and how AI is involved in their journey
Industry Examples in MEA
Banking and Financial Services
Banks in the GCC are deploying Agentic AI for proactive fraud prevention, instant credit scoring, and personalised wealth management. Instead of waiting for a customer to request a service, AI systems can assess their financial behaviour and suggest tailored investment options or spending alerts.
“Customer loyalty is no longer earned once — it’s earned every single interaction.”
– Wes Fagan, Regional SVP, Strategy & CDesO at Endava
Retail and E-Commerce
In the retail space, Agentic AI enables dynamic pricing, automated inventory management, and hyper-personalised product recommendations. A major UAE-based e-commerce player is piloting an AI that adjusts promotions in real-time based on browsing behaviour and competitor activity.
Government Services
In countries like Saudi Arabia, where Vision 2030 places digital experience at the core of public service delivery, Agentic AI is being used to streamline licensing, immigration, and healthcare processes.
“When AI can renew your licence before it expires, reschedule your medical appointment, and remind you of a tax deadline — all in one system — that’s when CX becomes citizen experience,” said Wes Fagan, Regional SVP, Strategy & CDesO at Endava.
Measuring CX in the Agentic Era
Traditional CX metrics like Net Promoter Score (NPS) and Customer Satisfaction (CSAT) still matter, but they’re no longer sufficient. With Agentic AI, speed, accuracy, and proactive engagement become critical performance indicators.
“We’re now measuring AI-driven interactions by how many problems they prevent, not just how quickly they resolve issues,” noted Ebrahim.
Companies in MEA are beginning to add metrics such as:
- Proactive Resolution Rate (PRR) – percentage of issues resolved before customer contact
- Autonomous Success Rate (ASR) – percentage of AI actions completed without human intervention
- Customer Effort Score (CES) – reduction in customer effort thanks to automation
Skills, Talent, and Organisational Readiness
Deploying Agentic AI is not just a technology challenge — it’s an organisational one. Companies must address skill gaps, re-architect workflows, and invest in change management.
“Agentic AI changes job roles,” said Sandeep. “You’re not replacing customer service agents; you’re evolving them into AI supervisors, escalation specialists, and empathy-driven problem solvers.”
In MEA, where tech talent shortages are already a concern, upskilling programs are becoming a priority. Leading organisations are partnering with universities, tech providers, and government initiatives to create AI-ready talent pipelines.
The Next Five Years in MEA
Looking ahead, Agentic AI will likely move from early adoption in large enterprises to widespread use across mid-sized companies and public institutions. Integration with the Internet of Things (IoT), 5G, and edge computing will further amplify its potential.
“The winners will be those who move quickly but responsibly,” concluded Basil Ayass. “In the race for CX excellence, speed without sensitivity will only get you so far.”
Closing: A New Era of Engagement
Agentic AI is more than a technology trend — it’s a paradigm shift in how enterprises think about customer engagement. In MEA, where cultural nuance, rapid digital adoption, and competitive markets intersect, its impact could be transformative.
But success will depend on balancing autonomy with oversight, personalisation with privacy, and innovation with trust. For leaders willing to embrace this balance, the rewards could redefine not just customer experience, but the entire business model.
