Employees are right to feel unsettled. As AI embeds itself deeper into enterprise workflows, the questions it raises aren’t just about efficiency — they’re about the fundamental structure of work itself. Which roles survive? Which skills actually matter? And who bears accountability when a machine acts on your behalf?

Reports from the International Labor Organization and the World Economic Forum highlight AI’s growing impact and potential to displace workers in routine and administrative roles, alongside a rise in new, more specialized, technology-driven jobs.

The push toward AI adoption is being shaped by both efficiency gains and broader strategic transformation, says Kurt Muehmel, Head of AI Strategy at Dataiku. He says it’s a shift towards “higher productivity through greater efficiency,” but stresses that organizations ultimately have discretion in how those gains are deployed.

When productivity rises, companies can choose between several paths: reducing costs while maintaining output, keeping inputs steady while increasing production, or scaling up both inputs and outputs to expand the business. 

This includes decisions about “human labor” as part of overall inputs—meaning AI may either reduce workforce needs in some areas or enable growth that creates demand for new roles and capabilities, says Muehmel.

Rather than simply doing existing work with fewer resources, companies may use AI to “fundamentally expand their business” and reshape the roles and skills required.

“At a macro level, it remains uncertain whether AI will lead to net job losses or gains. It could either suppress employment in some areas or have a massively positive effect on employment because the entire economy is just that much more productive,” adds Muehmel.

REDESIGNING ROLES

Sami Alshwairakh, Regional VP for Saudi Arabia at Fortinet, details how job roles are evolving as AI takes over routine tasks. He notes that his industry, cybersecurity, is among the most affected by AI, but in a positive way. 

He notes that cybersecurity has faced a skills shortage for years, with a global gap of about 4 million professionals.

“Instead of replacing any workers in our sector, AI helps overwhelmed security operations teams cope with the ever-increasing workload that is fuelled by malicious actors using AI-powered techniques to hit targets harder and faster.”

He explains that AI is establishing a need for new skills. “Finding cybersecurity talent with AI experience is the top recruiting challenge, 92% of organizations are likely to invest in AI-related cybersecurity training or certifications.”

At the same time, half of IT leaders find it hard to get approval to hire more cybersecurity staff, even though awareness of risks and regulatory pressures is growing.

Emile Abou Saleh, Vice President, Emerging Markets at Proofpoint, says AI is reshaping roles by fundamentally changing how employees interact with systems, data, and collaboration platforms. As organizations introduce AI assistants and autonomous agents, more responsibility is being delegated to systems that can draft communications, access sensitive information, and take action at machine speed.

“This means employees are increasingly working in environments where human users and AI agents operate side by side across email, cloud, and collaboration tools,” she states. “Job responsibilities are evolving to reflect that, because governing an AI agent that acts on your behalf requires a different kind of accountability than managing your own access.”

She adds that this convergence is also reshaping security and risk frameworks. “The bigger shift is that AI agents and human users now carry similar risks; both can be manipulated, and both can act outside their intended purpose.”

IMPACTED INDUSTRIES

Muehmel says AI is affecting industries in waves, depending on when new technology becomes available and is widely used.

“There is typically a lag of one year, a year and a half, or two years between the release of new AI systems and their integration into enterprise workflows. As a result, what we are seeing today reflects the impact of models released a couple of years ago, which were particularly strong at handling language.”

This is why jobs focused on language, like journalism, legal work, translation, and some customer support roles, are among the first to see major changes.

“Customer support is another area seeing early impact, particularly as speech-to-text and real-time conversational systems improve. However, this is only the first phase. The next wave of disruption is expected to come from AI agents that can ‘think through a problem… and then, of course, take action’, shifting AI from content generation to workflow execution,” says Muehmel.

As this develops, the impact is likely to extend to more process-heavy industries, such as financial services, manufacturing, and pharmaceuticals. These sectors involve structured decision-making steps—collecting data, analyzing it, and making informed decisions—which can increasingly be represented as defined processes that AI systems can support or execute. 

Abou Saleh adds that sectors with high data volumes and fast-moving communication environments, such as financial services, professional services, and technology, are likely to feel the shift earliest. 

Still, she notes that many organizations are struggling to keep up with governance and security. “Our research shows that in the UAE, 92% of organizations have already moved beyond piloting AI, yet 55% still describe their security posture as catching up, inconsistent, or reactive. That gap becomes much harder to manage as AI scales across already complex environments.”

THE UPSKILLING FACADE

As employees worry more about losing their jobs, many look for ways to keep their positions and stay ahead of automated systems that could replace them.

Workers often view upskilling as a potential solution, but Muehmel says, “Simply learning to use tools like ChatGPT or Claude more effectively is not, in itself, sufficient to address how work is being reshaped.”

“Instead of focusing narrowly on upskilling, the shift is toward thinking about what it means to do a job or to be an expert in some domain,” he adds.

In many workplaces, such as manufacturing or complex industrial settings, the most valuable skill is not just doing tasks more efficiently but using deep expertise to rethink how work is organized.

For example, rather than improving individual task execution, the role of an experienced operations manager increasingly becomes about “thinking about how the job should be done” and translating that into more efficient, AI-augmented processes.

“This reflects a broader shift from task-based work to process-based thinking. The most durable human value lies in understanding complex, organization-specific knowledge that AI systems do not inherently possess,” says Muehmel.

He notes that while AI may recognize general patterns in public data, it does not know “what it means to operate a specific manufacturing plant,” meaning human expertise remains essential for adapting systems to real-world contexts.

“In this framing, employees are not simply becoming operators of AI tools, but designers of workflows—translating tacit, experience-based knowledge into systems that combine human judgment and machine capability.”

He adds that this extends to fields such as medicine, where even if AI can outperform humans in narrow tasks like image detection, “accountability is the uniquely human task”, and broader responsibilities like patient care and treatment design remain inherently human. 

“As a result, the focus shifts away from simple upskilling toward a deeper redefinition of roles as ongoing processes rather than fixed sets of tasks.”

ETHICAL CONSIDERATIONS

Alshwairakh believes CEOs play a major role in setting priorities and strategies for integrating AI. “We see that success is not about ‘having AI’, but how it is integrated into processes. AI can only be truly valuable if it is embedded into core workflows and systems versus bolt-on solutions.”

He says that leaders and organizations should continue to tap into underutilized talent pools and invest in training and upskilling to build and retain the expertise they need. 

“This requires a coordinated approach grounded in three key pillars: raising awareness and education, expanding access to targeted training and certification, and deploying advanced security technologies.”

Abou Saleh says accountability has to come first. 

“When an AI system accesses sensitive data, drafts a communication, or carries out a task on someone’s behalf, there needs to be clear ownership of that outcome. As organizations redesign roles around AI, defining who is responsible for what becomes a basic ethical requirement.”

She adds that transparency matters just as much. Employees working alongside AI systems should have a clear understanding of how those systems are being used, what they can access, and where the boundaries sit.

“AI is also accelerating existing challenges around data handling, access, and decision-making at a much larger scale. Governance and accountability, therefore, need to evolve at the same pace as adoption, especially as these systems become more embedded into everyday workflows.”

THE IDEAL COLLABORATION

An ideal collaboration will be one in which people work with AI, moving away from constant task execution and focusing more on reflection and process design. Instead of always doing tasks, employees would spend more time thinking about what needs to be done, the main goals, and how to improve or redesign complex work.

“In this model, AI handles more of the routine or structured execution work… The human role becomes less about continuous task completion and more about stepping back to evaluate goals, priorities, and methods of execution at a systems level,” says Muehmel.

This change will not happen automatically or in the same way for every job. Muehmel says some roles will still focus on execution, but many will shift toward a clearer split: AI handles set processes, while people focus on judgment, values, and oversight. This includes making decisions based on human, ethical, moral, and social values, and using expertise that AI cannot match.

“Ultimately, the goal is not to eliminate human work, but to change its nature—shifting from task-level execution to process-level thinking, where humans work alongside AI to define, refine, and continuously improve how work is done.”

Alshwairakh says that going forward, AI governance is not a side issue but a top business priority. Careful oversight helps organizations benefit from AI while keeping trust, compliance, and resilience in a fast-changing digital world.

Also, using AI tools is only part of the solution. Success depends on preparing people to work with AI. Leading organizations are already investing in training programs that build AI skills for both technical and non-technical roles. These programs help employees spot bias, question results, and trust AI when used responsibly.

“Employees need AI literacy to understand outputs, recognize limitations, and maintain critical oversight – the human in the loop,” adds Alshwairakh.