57% of US firms now use AI in hiring, most often to review CVs

Once you send off your CV, it may be reviewed by AI first. A Resume.org poll reported by HCA Magazine (HRD America) on 25 August 2025 found that 57% of US companies already use AI in hiring, and that reviewing CVs is its most common use, at 79%. A further 74% of companies plan to use it more over the next 12 months.

What matters most to job seekers is who has the power to say no. The same poll found that 35% of companies reject candidates on AI recommendations at any stage, 39% do so only at the initial stage, and 26% never let AI take part in rejections. In other words, the first two groups add up to 74%, so most surveyed companies reject candidates on AI recommendations at some point in the process.

Applications have doubled, and recruiters are expanding their use of AI

LinkedIn published its research Talent 2026 on 7 January 2026. According to HR Dive's report the next day, 93% of the recruiting professionals surveyed plan to expand their use of AI in 2026, to meet hiring goals, assess candidates and find talent. 66% intend to use AI more for pre-screening interviews, and 59% already use AI to find overlooked talent.

Pressure lies behind the figures. In the same research, 42% of recruiters have been asked to fill roles faster, 66% say finding quality talent has become harder, and 73% feel unprepared for this year's workload pressure. HR Dive also noted that the number of applicants per US vacancy has doubled since spring 2022.

Other surveys point the same way. HR Dive cited a report published in October by Employ Inc., which said 65% of recruiters already use AI tools. HCA cited HireVue as saying that 72% of HR leaders worldwide use AI weekly, most commonly for CV screening and communicating with candidates. Greenhouse's 2026 AI in Hiring Report, published on 24 February 2026, surveyed 1,200 US job seekers, 219 recruiters and 446 hiring managers, and likewise said recruiters rely on automation to cope with large volumes of applications.

On the job seekers' side, trust is draining away

Greenhouse published its 2026 Candidate AI Interview Report on 1 May 2026, surveying 2,950 job seekers, of whom 1,200 were in the US and the rest in the UK, Germany, Australia and Ireland. It found that 63% had taken an AI interview, 13 percentage points more than six months earlier.

The data, drawn mainly from US respondents, deserve a closer look: 70% were not clearly told that AI would assess them; 38% had dropped out of a hiring process because of an AI interview; only 21% believe most employers use AI responsibly; and 57% think disclosing AI use should be a legal requirement.

Among those who completed an AI interview, 51% never received any reply. Only 28% moved forward, and 13% were rejected.

HCA also cited Gartner as saying that a quarter of job seekers would trust an employer less if it used AI to assess their data. Greenhouse's February report described job seekers as feeling they must "optimise" themselves to stay competitive, and as seeing transparency as the key to building trust.

Before reading the numbers, know their limits

Most of the data above come from the firms' own surveys by recruitment platforms or related vendors, including LinkedIn, Greenhouse, Resume.org and HireVue, and are not independent academic research. Some of the reports do not give sample sizes or survey dates: HR Dive did not list the sample size of the LinkedIn survey, Employ's report gives only October with no year, Greenhouse's job seeker survey page does not state when the survey was run, and the Resume.org and Employ figures reach us only through media reports. Treat these figures as a guide to direction, not as precise proportions.

Five ways to write your CV to reduce the risk of being wrongly screened out

The following is general advice. No survey shows that any CV format can guarantee getting past AI screening; the aim is simply to make sure both machines and people can read you.

1. Keep the format simple: list work experience, education and skills under clear section headings, and put key information in text rather than only in images or a complex layout.
2. Match keywords to real experience: use the wording of the job description to describe work you have actually done, and do not pad with skills you lack, because a human interview will eventually ask about them.
3. Give specifics for each role: stating your scope of responsibility, the tools you used and verifiable results is more convincing than adjectives.
4. Keep a route to a real person: besides submitting the application, follow up directly through the company's public contact details or someone you know. In the Greenhouse survey, the protections job seekers most wanted were a choice of a human interview (46%), disclosure in advance (44%), an explanation of what the AI measures (39%) and human review of AI assessments (38%); if you face an AI interview, you can ask the employer about these.
5. Set your own follow-up date: since more than half of those who completed an AI interview never received a reply, rather than waiting passively, note the date of each application and enquire once when it falls due.

Employers use AI to cope with volume. What job seekers can do is make sure the same CV tells the same story to a machine and to a person, with every line able to withstand questioning.

Prices and details correct as of November 2026

Compiled by the F50 Editorial Team from public sources, for general reference only and not professional advice. Details are correct as of the publication date and may change.