Researchers at the University of East London’s Royal Docks School of Business and Law conducted a systematic review of 79 peer-reviewed studies spanning computer science, organizational psychology, and human resource management. Their analysis reveals that AI shines in routine, high-volume tasks. Natural language processing algorithms parse résumés for keywords, machine learning classifiers match applicant profiles to job descriptions with remarkable accuracy, and automated scheduling systems eliminate the tedious back-and-forth of coordinating interview times. These tools compress weeks of administrative labour into hours, slashing time-to-hire metrics that corporate boards love. But the study’s deeper finding is that recruitment is not merely a logistical challenge; it is a social ritual of mutual evaluation, and candidates are acutely sensitive to how that ritual is mediated.

The core tension of the resourcing paradox hinges on what engineers call the black-box problem. Many state-of-the-art models—deep neural networks trained on historical hiring data—make decisions that even their creators cannot fully explain. A candidate rejected by such a system rarely receives meaningful feedback, breeding perceptions of unfairness. When organizations fail to disclose where algorithms are used, trust erodes further. Technically, this could be mitigated through explainable AI (XAI) methods such as LIME or SHAP, which highlight the features that influenced a particular decision. However, the review found that most employers are not deploying these transparency tools, leaving applicants to wonder if an invisible, biased logic gate ruled them out. In a hyper-competitive talent market, that suspicion is enough to drive sought-after engineers, designers, and leaders toward firms that feel more human.

Professor Kirk Chang, a co-author of the study, frames the issue squarely: “The organizations that win the competition for talent won’t be those that use the most AI. They’ll be the ones that combine AI’s speed with human judgement, transparency and empathy.” His point touches on a well-known limitation of current AI: it struggles to assess the kinds of latent qualities that define high-impact hires—empathy, creativity, leadership potential, and cultural contribution beyond mere culture-fit. While sentiment analysis of video interviews attempts to gauge soft skills, its validity is contested, and the algorithms can be confounded by neurodivergent communication styles or cross-cultural expression. A skilled human recruiter remains far better at reading the contextual nuances that signal genuine potential.

Co-author Professor Toyin Adisa underscores that recruitment often serves as a candidate’s first authentic encounter with an organization’s character. If that encounter feels impersonal—an endless series of uploads and auto-replies, or a final rejection from an unaccountable system—companies hemorrhage goodwill. The review cites evidence that candidates who perceive an automated selection process as unfair are less likely to accept offers, recommend the employer, or even remain customers. From a technical standpoint, the study calls for rigorous auditing of AI recruitment tools for both statistical and societal bias. This means testing models not just for overall accuracy but for disparities across protected groups using fairness metrics such as demographic parity or equalized odds, then applying bias-mitigation algorithms at pre-processing, in-processing, or post-processing stages. Few organizations currently perform such audits consistently.

The systematic review also highlights a critical blind spot: AI-driven recruitment systems are often optimized solely for efficiency, a single-objective function that ignores candidate experience. A scheduling bot that reduces recruiter workload by 80 percent is celebrated, but if it fails to answer a candidate’s nuanced question or routes her to the wrong interviewer, the damage to employer brand is measurable. The researchers advocate for a multi-objective optimization framework where candidate satisfaction scores carry weight alongside throughput metrics—a technically feasible but culturally underadopted practice.

The implications of the resourcing paradox extend far beyond HR departments. As AI regulation matures, most notably with the European Union’s AI Act classifying hiring algorithms as high-risk, companies that do not embed transparency and human oversight into their systems may face legal sanctions as well as talent defections. The study’s recommendations—disclose AI use plainly, maintain meaningful human review at crucial decision gates, and continuously audit for bias—are both ethical guardrails and strategic imperatives. In a world where the best talent increasingly chooses employers like consumers choose brands, the firms that master the delicate interplay between algorithmic efficiency and authentic human warmth will define the next era of work.

Subject of Research: The impact of AI-powered recruiting on organizational attractiveness, efficiency, and the “resourcing paradox”
Article Title: The resourcing paradox: a systematic review of efficiency and effectiveness in AI-powered recruiting
News Publication Date: 22-Jun-2026
Web References: https://www.emerald.com/er/article-abstract/doi/10.1108/ER-05-2025-0337/1382886/The-resourcing-paradox-a-systematic-review-of?redirectedFrom=fulltext
References: 10.1108/ER-05-2025-0337
Keywords: AI recruiting, resourcing paradox, efficiency, algorithmic transparency, candidate experience, human oversight, algorithmic bias, explainable AI