AI keeping us honest

Sounding polished is the part of AI hiring that gets the headlines. Candidates use it to write like professional writers, and companies use it to write job ads. That can tilt a screening model toward a certain style. It is a small fib next to the ones that change whether a job is real, or whether the person can do it.

In a December 2025 Kickresume survey of its own users, 55% said they use AI to make their CV sound more confident and professional. Xu, Li and Jiang held the facts on 2,245 résumés fixed and rewrote only the summary. Larger models still favored their own version over the human one by a gap of 67 to 82 percentage points. That is a preference for style, not for correct spelling, and not for honesty. The model was not asked who was telling the truth. It was asked which summary it liked.

The larger fibs run both ways.

On the applicant side, a May 2025 AIResumeBuilder poll of 7,884 US adults found that 10% of job seekers admitted lying on a résumé, and 31% of those said they used AI to help write it. Among Kickresume's users, 4% said they had made up or inflated titles, dates or qualifications. The common request was to sound better. Inventing the work was rarer, or at least more rarely admitted.

On the employer side, Resume Genius surveyed 1,500 US hiring managers in June 2026. 64% said their company had misrepresented some aspect of a role or the hiring process, most often by understating how hard the job or the team would be, or by giving an inaccurate timeline. 90% believe companies intentionally post job listings they do not plan to fill immediately. The listing and the opening, in that telling, are not always the same thing. 46% said their company uses AI to write job descriptions. The ads, like the résumés, are getting easier to make smooth.

In a June 2025 Kelly survey of more than 1,000 hiring managers, 81% had posted a job, run a full interview process, and then decided not to fill it, and 54% admitted misrepresenting remote work availability to secure hires. Stepstone and HTW Berlin found that 71% of HR decision-makers say job ads often ask for more qualifications than the role actually needs. The post underneath those requirements is a separate question.

AI will not sit across the table and know who is lying. It is not a lie detector. What it will do is the homework that busy people skip. It can look up whether a company, a degree, a paper, or a job title leaves a public trace. It can notice when dates do not line up, when a résumé's claims do not match a work history, or when a posting's remote line does not match the career page. Those are cues, not verdicts, and they show up on both sides of the ad.

Vendors already sell some of that lookup work. What employers actually use AI for is still mostly scoring writing. In HireRight's 2026 survey, 17% of North American respondents used AI for candidate selection or résumé filtering, the most common HR use, and 52% used no AI in HR at all.

The interesting use of AI in hiring is not making everyone sound like a copywriter, and it is not catching every lie. It is doing the homework and picking up cues from the applicant and from the company. We still do not have a measure of whether résumés or job posts got more or less true after generative AI arrived. We do have both sides saying the larger fibs are already common.

The figures live on Hiring: both sides use AI.