UK employers are deploying AI across the full recruitment stack, from CV screening and automated scheduling to video interview analysis, while candidate preference data creates a genuine tension for HR leaders designing those systems. Roughly 70% of British workers still prefer face-to-face interviews, even as companies like Capita report a 43% reduction in time-to-hire after adopting AI-supported hiring. The AI recruitment shift<\/a> is accelerating demand for AI governance professionals and HR technology specialists alongside the expected engineering roles, which means the function is both adopting the tools and being reshaped by them simultaneously.<\/p>\n The gap between what employers are automating and what candidates will tolerate is not a soft cultural concern. It is a sourcing risk. Organizations that automate early-stage screening without transparency about how decisions are made will see candidate drop-off and, given the UK Equality and Human Rights Commission’s existing warnings on discriminatory automated decisions, regulatory exposure. CHROs whose AI stack is vendor-assembled rather than intentionally designed are most exposed here, because the liability sits with the employer, not the platform provider.<\/p>\n The Amazon case in the article deserves more weight than it gets. Amazon’s automated hiring system had to be scrapped after it systematically disadvantaged women, and that outcome happened inside one of the world’s most sophisticated ML organizations. The implication is not that smaller enterprises should avoid AI recruitment tools. It is that organizations without dedicated AI governance capacity, someone accountable for auditing model outputs against protected characteristics, are running a system they cannot actually see. The Unilever and Capita examples look cleaner partly because both invested in human oversight at the decision stage, not just at the screening stage.<\/p>\n The 2025 research finding that candidates using AI-generated answers scored higher on performance metrics but triggered fairness concerns points at a structural problem heading toward every CHRO’s inbox. If AI helps candidates perform better on AI-evaluated interviews, the system is measuring preparation for the tool rather than job-relevant capability. The budget question this reframes is not whether to buy an AI recruitment platform. It is whether the evaluation criteria those platforms optimize for have been validated against actual job performance in your organization, because if they haven’t, you are paying to hire people who are good at interviewing AI.<\/p>\n An asynchronous video interview is a recorded response format where candidates answer preset questions on their own time, with no live interviewer present. AI systems then score the recordings by analyzing verbal content, speech patterns, and response structure. Think of it as a standardized written test, except the medium is video. For HR leaders, the operational appeal is obvious: one set of questions, consistent scoring, no scheduling overhead. The governance risk is that the scoring model’s assumptions about what “good” looks like may not reflect your actual top performers.<\/p>\nWhat this means for your business<\/h2>\n
Concept deep-dive: Asynchronous Video Interviews (AVIs)<\/h2>\n