By Juliet Umeh

Olasunkanmi Lawal, Co-founder of UseHirable, shares his thoughts on the challenges of AI-driven recruitment.

Artificial intelligence entered hiring with a promise: faster decisions, fairer processes, and better alignment between people and jobs. For many job seekers, however, AI has yet to deliver. Instead, it often acts as an invisible wall. Applications vanish into systems that provide no feedback, explanation, or indication that a human ever reviewed them. The problem is not AI itself. It is that recruitment technology has been designed primarily for organisational efficiency, yet presented as neutral.

Most modern hiring systems aim to manage scale. Automated tools filter applications, shortlist candidates, and weed out CVs long before a human sees them. For organisations, this is efficient. For candidates, it can feel like being erased. Qualified applicants are left in the dark, unsure whether their experience mattered at all. The impact is particularly harsh on early-career professionals, career switchers, and international applicants. Rules treated as “neutral” by automated systems often filter out capable candidates, even when their skills and potential are strong.

Many believe algorithms are less biased than humans. In reality, most hiring systems simply reproduce the patterns on which they are trained. If past hiring decisions favoured certain schools, career paths, or demographics, those biases are embedded in automated filters. Bias does not disappear, it becomes quieter and harder to challenge. One of the least discussed harms of AI-driven hiring is silence. Candidates rarely learn why they were rejected or how to improve. What begins as efficiency quickly turns into detachment, and when decision-making systems control access to work, that detachment carries real consequences for stability and dignity.

If AI is to be used responsibly, it must be deliberate. Eligibility criteria should be clear, rejections should include explanations, and success should be measured by fairness as well as speed. Governance and accountability must be treated as design choices, not legal afterthoughts. Around the world, regulators are questioning the impact of opaque algorithmic decisions and public distrust is growing for good reason.

In countries like Nigeria, where unemployment and underemployment are persistent, the effects of opaque systems are felt more sharply. Frustration can turn into disengagement, and confidence in employers gradually diminishes. AI does not need to screen out applicants to be useful. More automation alone will not create progress. Real improvement comes from asking better questions: who are these systems designed to serve, who bears the consequences, and how can technology respect the human behind every application?

Hiring technology now shapes access to opportunity. With that power comes responsibility. If ignored, a tool meant to open doors risks quietly closing them.