In the race to lead the AI economy, a new kind of machine intelligence is stepping onto the stage: agentic AI–systems that don’t just follow instructions, but set goals, make decisions, and execute tasks independently. If traditional AI were a smart assistant, agentic AI would be the ambitious colleague who takes the brief, runs with it, and doesn’t wait for you to check in.

The narrative around agentic AI often leans toward fear: machines taking over, roles disappearing, and workers left behind. But history shows us that technological revolutions don’t just eliminate jobs—they also create entirely new ones.

Governments and businesses in the Gulf are betting big. The UAE’s AI Strategy 2031 and Saudi Arabia’s Vision 2030 treat AI not as a side project, but as a national growth engine. Deloitte Middle East has already rolled out an Agentic Network for clients in sectors from energy to finance. At this year’s Customer Experience Live Show Middle East, 89% of enterprises said they’re investing in predictive AI tools that anticipate and act on customer needs.

For workers, that ambition cuts both ways. On one hand, there’s a clear risk: routine jobs—think data entry, basic scheduling, and repetitive service roles—are the first in line for automation. In finance alone, AI is already automating up to 75% of an analyst’s tasks. On the other hand, there’s an opening for new careers: AI ethicists, prompt engineers, and“collaboration managers” who specialize in working with autonomous systems.

While there are concerns that agentic AI could hollow out middle-tier jobs, it could also help the region leap into a new high-skill employment era.

Since change is inevitable, sooner rather than later, can organizations train enough people fast enough to work alongside machines that think for themselves?

“It’s not going to be clear what the ’safe zones’ around AI will be, especially given how rapidly the technology evolves. Roles and functions requiring high trust, cultural nuance, and high-stakes accountability will likely move more slowly due to the associated risks of incorporating AI into these industries,” says Richard Johnson, COO, Data Guardians Network.

Some industries will be more affected than others. “Agentic AI is most likely to affect industries where high-volume decision making and problem-solving intersect,” says Christopher Crecelius, co-founder and CEO of AxonDAO.

Fields like logistics and fintech are primed for agentic AI use, requiring constant decision-making, data streams, and precise outcomes. In finance, it can strongly model market risk and adjust strategies in real-time without constant human prompting. “These are areas with massive data flows, clear objectives, and high rewards for speed,” Crecelius adds.

“Customer service, cybersecurity, and finance are seeing the earliest disruption because they run on repetitive workflows and data analysis – exactly what agentic AI excels at,” says Zac Cheah, co-founder at Pundi AI.

THE JOBS AGENTIC CAN’T TOUCH YET

While agentic AI is proving it can outpace humans in crunching numbers, processing paperwork, and running logistics, there’s one area it still can’t replicate: the human edge.

Experts say that, depending on emotional intelligence, cultural nuance, high-stakes judgment, or deep relationship-building—from senior negotiators and creative directors to healthcare specialists and policy advisors—remains firmly in human hands. Even in the most AI-forward economies, these “judgment-heavy” roles face far lower automation risk than routine administrative work.

Citing an example of healthcare R&D, Crecelius says agentic AI may be able to run simulations, analyze vast biological datasets, and generate hypotheses at striking speed, but it can’t design novel experiments, interpret anomalies in context, peer review, build theories, or weigh ethical considerations the way humans can.

“These things require intuition and creativity, something an algorithm can’t replicate. Science isn’t just data processing, but it’s the art of asking the right questions and navigating uncertainty,” he adds.

While there’s no guarantee of “safe jobs”, Johnson says there will always be a need for a ‘human-in-the-loop, ‘ meaning that industries won’t become redundant but will shrink the number of workers. “The rise of this new tool will create industries to support this technology’s growth,” says Johnson.

He adds that AI training, specifically in data labelling and frontier data creation, will become increasingly important to ensure these agents remain fit for purpose, accurate, and unbiased.

According to Cheah, agentic AI could create net employment gains in different areas. “We’re already seeing growth in AI development, human-AI collaboration roles, and education to help workers adapt.”

“The key preparation measure is investing in continuous, AI-powered reskilling that adapts to each worker’s needs, making learning ongoing rather than a one-time event while restructuring jobs so humans focus on what they do best: creativity and decision-making,” adds Cheah.

NEW JOBS ARE COMING. IF YOU’RE READY FOR THEM

Agentic AI is already opening doors to roles that didn’t exist a decade ago. While some roles may disappear, others—often more strategic, creative, or oversight-focused–are just beginning to take shape.

“Agentic AI will create roles such as AI oversight, AI-driven R&D, ethical governance, and human-AI collaboration,” Crecelius says, adding that previous waves of automation created opportunities rather than simply displacing workers.

He points to early examples: growing demand for AI systems integrators, who embed AI into complex operational systems, and validation specialists, who ensure AI-generated work meets legal or compliance standards. Crecelius adds that creative industries create new jobs around curating, directing, and refining AI outputs.

At the same time, he warns that companies and individuals who ignore agentic AI risk job insecurity as the technology becomes more integrated into everyday business operations.

THE NEW SAFETY NET FOR AN AGENTIC AI WORKPLACE

Preparing for an agentic AI future isn’t just about upskilling; it’s also about protecting the people who work with these systems. Crecelius says there are many steps businesses and independent workers can take. Ensuring anyone using agentic AI has tailored insurance coverage that protects against errors, bias, or unintended outcomes.

“Similar to how cybersecurity insurance became essential in the digital age, AI liability coverage will be non-negotiable for managing operational and reputational risks,” he says. This coverage would safeguard companies and workers if an AI system makes a costly or damaging mistake. Crecelius adds that pairing liability protection with ongoing AI education and weaving both into company culture will empower employees to carry out tasks confidently, knowing both skills and a safety net backs them.