The AI will see you now
AI is rewriting the HR rulebook, changing everything from recruitment through to exit. But this innovation is not without its risks.
Artificial Intelligence (AI) is rapidly transforming the human resources (HR) landscape, leading to significant job cuts across large technology firms. As of mid-July 2026, nearly 140,000 jobs had been shed in the sector, with companies like Cisco, Oracle, and Thomson Reuters announcing workforce reductions due to AI adoption. South Africa, meanwhile, has yet to experience mass corporate layoffs directly linked to AI.
However, AI is proving beneficial for HR in various ways. It assists in streamlining hiring processes, answering employee queries, summarizing performance reviews, and onboarding new staff. According to research, AI is expected to increase productivity by 35%, improve training effectiveness by 30%, and boost retention by 20% by 2027. Moreover, AI can help businesses forecast future skills needs by analyzing workforce data.
HR professionals estimate that up to 60% of their time is currently spent on basic queries, such as checking leave balances or payroll information. By deploying AI assistants to handle these routine tasks, HR teams can reclaim valuable time. For instance, AI can help draft job descriptions, compile interview questions, and personalize candidate communications. This not only saves time but also allows HR teams to focus on more strategic tasks like workforce insights, skills analysis, and workforce trends.
Despite these benefits, there are still concerns about AI's impact on HR, particularly regarding governance, accountability, privacy, potential bias, trust, and compliance. AI agents that can autonomously carry out certain HR workflows require complete transparency in decision-making to ensure trustworthy outcomes. It is crucial to strike a balance between leveraging AI's potential and maintaining human judgment in complex organizational dynamics, change management, and employee trust-building.
Written by urgent.news from ITWeb's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
