AI in Recruitment 2026: The State of Play
If you are applying for jobs in 2026, you are almost certainly using AI. So is the person screening you. And here’s where it gets a little bit contradictory, even slightly hypocritical. The organisation asking whether you used AI on your cover letter is probably using AI to decide whether a human ever reads it.
Before you can use AI well in a job search, you need an honest picture of the market you are applying to. So here is the state of play on AI in recruitment, drawn from the current UK and international data, and what it means for how you should apply.
Everyone is using it, on both sides of the table
Start with the candidates. Greenhouse’s multi-market research across the US, UK, Ireland and Germany found that 78% of European candidates are using AI tools in their job search, with nearly half saying they do it specifically to get past automated filters. In the UK, Omni’s February 2026 survey of 739 candidates put AI use in applications at 47%, and Totaljobs research reported in May 2026 found recruiters estimate around 41% of the applications landing in their inboxes have been AI-enhanced. Among graduates, the figure is higher still: a January 2026 Kahoot survey found 65% using AI to assist with applications.
Now the employers. CV-Library’s March 2026 research found 83% of UK recruiters use AI somewhere in their hiring process, with 28% relying on it to screen applications. Greenhouse found that 61% of US hiring managers, and 59% across the UK, Ireland and Germany, now run detection software to spot when candidates have used AI.
Read those two paragraphs together, and the picture is clear. This is an arms race. Candidates use AI to beat the filters; employers use AI to run the filters and then more AI to detect the candidates using AI. Nobody planned this. It is simply what happens when both sides adopt the same technology and neither side talks about it. I’m in a WhatsApp group and a similar circle of activity played out around how advice is given by large consultancies and large organisations currently too. But that is for another day.
140 applications per graduate vacancy. That is the current UK average according to the Institute of Student Employers, a historic high that has now held for two consecutive years. In 2002/03 the figure was 38. In the most competitive sectors it approaches 300.

I graduated in 2001 with my BSc and then again with my MSc in 2002. I was in the unlicky 37. I’m not bitter. It was hard then. It’s off the scale in terms of difficulty today
Volume is up, signal is down
The ISE numbers above explain why. AI and one-click apply tools have collapsed the cost of submitting an application, so volumes have surged: ISE members received over 1.8 million applications for around 31,000 early-careers roles in the latest survey year. Employers respond to volume the only way they can at that scale, with automated screening. Which pushes candidates to optimise for the machines. Which increases volume further.
The result is that trust has become the scarcest commodity in hiring. Greenhouse reports 91% of recruiters have spotted candidate deception of some form, while more than half of candidates suspect AI has evaluated their applications without being told. In CV-Library’s research, 53% of UK jobseekers believe they have been rejected by AI without a human ever reading their CV, and 63% say AI-led hiring feels less fair than human-led hiring.
The AI interview has gone mainstream
The most striking shift in the past year is not screening, it is interviewing. Greenhouse’s 2026 Candidate AI Interview Report, surveying nearly 3,000 active job seekers, found that 63% have now been interviewed by an AI, up 13 percentage points in just six months. Transparency has not kept pace: 70% were never clearly told upfront that AI would be evaluating them, and one in five only realised once the interview started.
Candidates are voting with their feet. 38% have walked away from a hiring process because it included an AI interview. Of those who completed one, half never heard back at all. Whatever you think of AI interviews, they are no longer an edge case you might encounter; they are part of the standard process you should prepare for.
Hypocrisy and the double standard
Here is the part that should shape your strategy most. UK employers hold candidates to a standard they do not apply to themselves. Research reported in May 2026 found just 16% of UK HR decision-makers view candidates using AI positively, and 28% view it negatively, more than double the global average. The same market in which 83% of recruiters use AI themselves.
I am not pointing this out to fuel or encourage cynicism. I am pointing it out because it tells you exactly where the line sits. Employers are not objecting to AI; they use it constantly. They are objecting to AI-generated candidates: applications where the tool has replaced the person, where the CV overstates, where the interview answers do not match the cover letter. Kahoot’s survey found 51% of graduates admit to exaggerating their skills when using AI, and that is precisely the behaviour the detection software and the sceptical hiring manager are hunting for.
The rule that falls out of all this data: AI that amplifies your genuine evidence gets you through. AI that substitutes for it gets you filtered.
The regulator has arrived
The other development that changes the landscape is regulatory. On 31 March 2026 the ICO published its report on automated decision-making in recruitment, alongside draft guidance, drawing on evidence from more than 30 UK employers. The headline finding was awkward for the industry: many employers told the ICO their AI tools were only used for decision support, with a human making the final call, and the ICO concluded that several were in fact making automated decisions without the safeguards the law requires. Sixteen named organisations received letters and have committed to act.
Why does this matter to you as a candidate? Because those safeguards are your rights. Where automated decision-making applies, you are entitled to know it is happening, to make representations, to request human review of a rejection, and to challenge the decision. Most candidates have no idea these rights exist. The ICO’s final guidance on automated decision-making is now expected in winter 2026, following the consultation that closed in May, and I will cover what it means for job seekers in a separate post outside of this series.
What this means for how you apply
Pull the threads together, and the strategy writes itself.
First, opting out of AI is not a strategy. When 140 people are applying for the role, and most are using AI, unassisted applications compete on volume terms they cannot win. The question is not whether to use AI but how.
Second, generic AI output is now actively harmful. Detection software is widespread, recruiters have seen thousands of identikit AI cover letters, and the market penalises anything that smells generated. The lazy use of AI that worked in 2024 gets you filtered in 2026.
Third, the winning position is the one the data keeps pointing at: real, specific, verifiable evidence of what you have done, presented clearly, with AI doing the structuring and sharpening rather than the inventing. That is the foundation of the Evidence-First approach I set out in the previous post in this series, and it is the approach the rest of this series builds on: decoding adverts, tailoring CVs, passing the authenticity test, and preparing for interviews, including the AI-led ones.
The market has changed faster than the advice. But the fundamentals have not: the person doing the hiring wants to meet a person. Use AI to make sure the person they meet is the best-evidenced version of you.
Get the full picture, and the system for navigating it
This post is part of a series drawn from my whitepaper, The Evidence-First Job Search: a complete framework for using AI across job search, applications and interviews, with a 15-prompt library, an evidence bank template, and a 30-day plan. It is free to download.

0 Comments