Can you hire AI for a job?

The National

27-05-2025

Artificial intelligence has been on a hot streak and, while it's a boon for some, it has been bane for jobs. Most of the attention has been drawn to the negatives – jobs displaced or wiped out – but studies have found the technology to be an opportunity to create more and reskill employees. Human resources departments, meanwhile, have also been trying to figure out how to balance tradition with the hot innovation. "The biggest change will be people working alongside AI, to capture the upside of 'augmented intelligence', so there will be a race to equip people with the right mindsets, skill sets and toolsets,' Lisa Lyons, regional transformation centre of excellence lead at New York-based professional services firm Mercer, tells The National. "Another bright side, is making work more interesting … employees report that their work today is mundane and repetitive, presenting an obvious opportunity for process and cognitive automation.' Analysts have drawn parallels to other industrial revolutions, in which the job market was disrupted, in what is considered its fourth iteration. The big difference is that the technology has set the pace faster, and everything else must keep in step. And while it is universally agreed upon that AI will indeed replace or displace jobs, figures have varied. But, in fact, jobs continue to be created as eras open up roles and demand. The International Monetary Fund estimates that 60 per cent of jobs in advanced economies would be affected by AI; that number drops to 40 per cent 26 per cent in emerging and low-income economies, respectively. "As AI continues to develop, it will become increasingly adept at performing a variety of roles that have traditionally been done by humans,' says Mohammed Alkhotani, a senior vice president at cloud services company Salesforce Middle East. For example, autonomous AI agents, built on natural language processing and powerful reasoning engines, can mimic human language – both written and spoken – and are ideally suited to handle a wide range of roles. "This particularly applies to customer services, where AI agents are already working alongside humans,' Mr Alkhotani tells The National. Customer services are forecast to be among the top declining roles through 2030, the World Economic Forum said in its Future of Jobs Report 2025. Multiple studies agree on what jobs are at risk from AI, or the jobs the functions of which can be mostly performed by AI in a more accurate and cost-efficient way. Among those roles that have been automated, or are at risk of being automated, are telemarketers, customer service representatives, manufacturing assembly line workers, proofreaders and translators. On the flipside, jobs that are less at risk – or are outright difficult to replace using the technology of today – are those that need people to provide emotional understanding, interpersonal skills, human judgment and adaptability, according to Kieran Gilmurray, an AI strategist who founded an IT services firm bearing his name in Northern Ireland. Those include jobs in health care (doctors, surgeons, nurses), the arts (sculptors, musicians), social workers and counsellors and skilled tradespeople (electricians, carpenters, plumbers). "AI should complement human expertise, not replace it … decision-making often requires nuance, negotiation and adaptability, which AI cannot fully replicate,' says Ibrahim Imam, Vienna-based construction software developer PlanRadar's chief executive for the Middle East and North Africa, and Asia-Pacific regions. "The key to successful AI integration is continuous training, contextual learning and a hybrid approach where AI handles data-heavy tasks, allowing professionals to focus on strategic, human-centric decisions.' Much of the focus – and worries – of the apparent job disruption has been towards the workforce, and HR units of companies are feeling the pinch. For instance, while HR teams are likely to incorporate more AI tools, the idea of AI fully replacing human HR professionals is highly unlikely for a number of reasons. While AI can streamline recruitment, onboarding, and employee performance tracking, it lacks the emotional intelligence needed for complex human interactions, says Nicki Wilson, managing director of Dubai-based recruitment firm Genie. "It also becomes a bit of a spambot in the sense that AI can often seek out job roles for jobseekers and send hundreds, if not thousands, of applications to adverts, decision makers and HR teams,' she tells The National. In addition, using AI tools to create CVs are "actually counter productive', as most traditional software that recruiters and employers use cannot read these CVs, which are effectively images, she says. "This, honestly, is not helping anyone trying to hire.' The WEF agrees. AI systems still largely rely on self-reported candidate information, making them susceptible to inaccuracies, the jobs report says. It added that around 88 per cent of companies have already used some form of AI for initial candidate screening. However, that has already been a trend, even before the pandemic year – and way before ChatGPT burst into the scene, a study from US-based HR services firm SHRM finds. "What's more, these systems can also filter out highly qualified, high-skill candidates if their profiles don't match the exact criteria specified in the job description,' the WEF report stresses. In the US, inefficient career transitions and learning gaps are costing the world's biggest economy about $1.1 trillion annually, research from UK education conglomerate Pearson shows. That underscores the need for more effective workforce development solutions, as AI's role in workforce development "extends far beyond improving efficiency', a representative from the London-based company tells The National. Jobs that require human interaction, creativity, critical thinking, strategic decision-making, emotional intelligence and advanced technical expertise are best positioned to thrive amid the AI boom, recruitment consultants had previously told The National. "AI-driven platforms can offer personalised learning experiences … these platforms analyse employees' individual learning styles, strengths and weaknesses to deliver content tailored to their needs, helping to bridge skill gaps effectively and efficiently,' the Pearson representative says. That does not mean companies should just roll with the punches. A study from the California-based non-profit Rand Corporation finds that more than 80 per cent of AI projects don't succeed. That "emphasises the need for clear guidelines and practical applications', says Evgenii Pavlov, general manager at Yango Ads Middle East and Africa. "The technology landscape is littered with instances where AI was applied unnecessarily, resulting in failures and unmet expectations,' he tells The National. The reliability of AI systems has always been under scrutiny. It is not uncommon for chatbots or – text-based or voice – which are increasingly replacing humans in call centres, to take users in circles. Some do not even have outright options to request for an actual person to speak to. The reason for this is issues such as limited understanding of complex queries, and poor training data, in addition to scaleability problems wherein bots still in development, with limited testing, have been rolled out, leading to inefficiencies. "Finally, the process of human-AI collaboration, such as escalating to a human agent, can sometimes be inadequately managed, leading to circular interactions,' says Louis Mottli, founder and chief executive of UK-based entertainment app developer Mottli. "At its core, AI has always been an enablement technology, designed not to replace but to enhance human thinking, decision-making and execution … businesses don't succeed by automating everything; they succeed by striking the right balance.'