Agentic AI is best suited for repetitive, data-heavy tasks such as preparing reports, monitoring performance and flagging unusual behaviour
The UAE is on its way to become the first government in the world to have 50 per cent of its services operated by artificial intelligence. Both the government and private sector will adopt agentic AI either through training staff members and public officials or by helping companies integrate AI into their daily operations.
In early May, the Dubai Crown Prince Sheikh Hamdan bin Mohammed bin Rashid Al Maktoum announced a new initiative to accelerate the adoption of agentic AI in Dubai’s private sector, while Sheikh Mohammad bin Rashid, the ruler of Dubai, announced shortly after that the UAE government will have 50 per cent of its services run by agentic AI, becoming the first country in the world to do so.
But what exactly is agentic AI, and how different is it from generative AI like Claude and ChatGPT?
What is agentic AI?
Agentic AI is semi- or fully autonomous, meaning it can make decisions on its own without having to be prompted every step of the way, explained Hetarth Patel, WebEngage’s Vice President of the Middle East and Africa, Americas, and Asia Pacific.
Whereas the AI most people use today requires prompts for it to take an action, agentic AI can take a goal and work towards it. “It figures out what needs to happen, talks to the right systems, executes, and flags anything that needs a human in the loop,” he said.
This is why agentic AI will be most useful for repetitive and data-heavy work, such as preparing reports, monitoring campaign performance, flagging unusual behaviour, and others. Because of its decision-making capabilities, agentic AI will be able to recommend the next best course of action when executing tasks, Patel explained.
At the same time, human involvement will remain critical, the VP added. “AI can recommend an action, but people must decide the rules under which it acts.”
How will employees be affected?
In the short term, agentic AI will reduce the load on employees, Patel said, adding that a lot of time goes into routine updates and follow-ups. “That’s where the immediate gains will show up,” he noted.
In the long run, some jobs will change shape as the nature of work shifts to less time executing every step and more time on other actions that require real judgement.
On the other hand, Vasudha Khandeparkar, an AI and data expert, admitted that some jobs may be replaced altogether. “If there are highly repeatable tasks that require low levels of input, judgement or contextual interpretation, those are very likely to be automated over time,” he said.
What could agentic AI look like in the future?
Nowadays, most people are already interacting with these technologies without even realising it, Khandeparkar said. “Think about the automated message you hear when calling a contact centre saying the call may be recorded for quality purposes. Historically, someone would manually listen to those calls, transcribe them, log outcomes and assess staff performance,” he said.
“Today, AI agents can transcribe the conversation automatically, assign quality and sentiment scores, identify customer frustration patterns and even determine which call centre staff may need additional coaching or training.”
This is just one example of what agentic AI could look like in the future. As the technology evolves, it is expected to take on many more autonomous and collaborative roles across different aspects of work and daily life.
