AI Writes The CV. AI Screens The CV. How Do We Make Sure We’re Still Hiring Humans?

Nearly half of UK job seekers use AI for job applications, whether that’s writing CVs, drafting cover letters, prepping for interviews, or even generating questions to ask the interviewer. On the other side of the table, HR.com reported how 37% of businesses avoid AI utilisation in the recruitment process, meaning 63% of employers actively use it.

In a world where AI is being used more by the day, we need to pause and ask: are we building a world in which algorithms increasingly talk to algorithms, with humans pushed to the edges of decisions that profoundly shapes people’s lives and the demographic of our workforces?

Companies are adopting AI at pace, drawn by the promise of efficiency at scale. But if we are not deliberate, AI will quietly reverse the progress made on inclusion. Not through malicious intent but through the way AI systems are designed, trained and deployed.

The risks hiding in plain sight

Most organisations are turning to AI to cope with soaring application volumes. The irony is that those volumes have been inflated by candidates themselves using AI to apply for more roles faster. Four shifts in particular carry real inclusion risk.

Keyword filtering and automated shortlisting

Most AI screening tools rank CVs by how closely they match keywords from the job description. Candidates who know this game mirror that language back, often using AI to do it for them. The result is a shortlist of nearly identical CVs, where the winners are those coached in ATS optimisation or with access to the right tools, not necessarily those best suited to the role or with transferable skills.

Reliance on referral hiring

Faced with application overload, many employers lean harder on referrals. But referral pipelines replicate existing networks, which in technology, financial services and senior leadership still skew towards white male candidates.

Standardised aptitude and logic testing

Depending on the usage and application, universal assessments sometimes only measure a narrow slice of capability – typically speed, pattern recognition and logical reasoning under time pressure. They can disadvantage candidates with different cognitive styles, neurodivergent profiles or cultural context, and they can miss the qualities that actually drive performance in the role: empathy, judgement, communication, resilience, the ability to build relationships or navigate ambiguity. A candidate who scores poorly on a timed logic test may be exactly the person a team needs to lead, manage a client, or hold a complex project together.

Asynchronous video pre-screening

Algorithmic analysis of eye contact, body language and “camera confidence” creates barriers for neurodivergent candidates and those from cultures with different communication norms. For instance, a candidate looking around the room, a common characteristic of neurodivergence, can be flagged as suspicious.

These biases aren’t new. AI is simply emphasising them and, in some cases, making them harder to see.

The race to adopt AI is uneven

A Harvard Business School analysis of 18 studies covering more than 143,000 people across 25 countries found women had 22% lower odds of using generative AI than men. Deloitte’s 2024 UK research found a similar pattern: 28% of women were using generative AI, compared to 43% of men. The gap was most pronounced among older workers.

Therefore, candidates from sectors and demographics already over-represented in hiring funnels are gaining an additional algorithmic edge, while older demographic, those returning from career breaks, candidates from lower socio-economic backgrounds and individuals writing in a second language, fall further behind. The “level playing field” AI was meant to provide is tilting in a familiar direction.

The good news: there are clear proactive actions that can be taken

AI has the potential to genuinely improve hiring quality and widen access, but only if inclusion is built into the design of the process from the start. Here are some key actions that will support building inclusion into the hiring process:

Human + AI symbiosis

Implement a process where the human recruiter and AI work together in a mutually beneficial relationship where each does what they do best. Use human recruiters to regulate AI outputs, to provide critical thinking throughout the process and keep inclusion checks in place.

Keyword filtering bias

Audit AI screening tools before deployment. Define role success by capabilities and outcomes, not keyword matches. Build skills-based assessments that surface transferable experience.

Over-reliance on referrals

Be aware of over-reliance on referrals. Referrals should be used as one channel among many. Audit referral practices for inclusion alignment and invest in outreach to underrepresented communities.

Assessment bias

Offer alternative assessment formats. Challenge what the role actually requires. The soft skills that psychometric tests miss may matter more than the ones they measure, especially in consultancy environments.

Video screening bias

Train interviewers in neurodiversity and cross-cultural communication. Avoid AI analysis of facial expression, tone or body language for selection. Use structured interviews evaluated against role-relevant criterias.

Across the process

Track diversity metrics at every stage of the funnel. Embed inclusion governance into AI procurement. Be transparent with candidates about where and how AI is used.

The opportunity

The organisations that build inclusive, human-led, AI-supported recruitment will win on every dimension that matters.

Whether organisations are hiring directly or with the support of recruitment agencies, they must confront the risks of AI from the offset in order to make strategic choices. Businesses can then use the approach as a market differentiator to build stronger teams and better talent pipelines, whilst gaining more innovation and a sustainable advantage in a market being reshaped by AI itself.

In a world where AI writes the CV and AI screens the CV, the competitive edge belongs to the organisations that remember they are still, fundamentally, hiring humans.

Written by Laura Barlow - CEO & Founder of Inclusive by Design Consulting

Laura is an award-winning leader, speaker and visionary in the tech industry, dedicated to reshaping workplaces into truly diverse, equitable, and high-performing environments.

Drawing on more than two decades in the tech industry, Laura challenges conventional approaches to workplace culture and advocates for systems that enable diverse talent to thrive. As the founder of Inclusive By Design Consulting, Laura works with organisations to turn equity from aspiration into action through practical, people-centred change.