Ghost Jobs, Ghost Candidates: Why Neither Side of Hiring Trusts the Other Anymore

For years, the hiring-trust conversation ran one direction: candidates complaining that job postings weren't real. Three reports released within the last two months of 2026 show that framing is now half the story. Employers have an equal and opposite complaint, and the data suggests the two problems are not separate — they're feeding each other.

What candidates are seeing

ResumeBuilder.com's Ghost Jobs Report, released October 14, 2026, surveyed 1,011 U.S. adults who had searched for a job in the prior 24 months (fieldwork: August 27, 2026). Eighty-six percent said they'd encountered a posting they suspected wasn't tied to an active hiring need — not "a company that's slow to fill a role," but a listing with nobody actually planning to fill it.

That figure sits inside a tighter, harder-edged estimate from the company that would actually know: Greenhouse, the applicant-tracking platform used by thousands of employers to run requisitions. Reporting on Greenhouse's own platform data (via the Wall Street Journal and corroborated by multiple outlets), Greenhouse classifies a posting as a "ghost job" when it received at least one application but produced no hiring activity — no test sent, no interview scheduled — for the quarter. Its own figure: 18–22% of postings on its platform in any given quarter, with Q2 2026 landing around 19%. That's not a subjective "felt fake" number. It's measured against the platform's own activity logs, which is about as close to ground truth as this question gets without subpoenaing an HR department.

Why do companies do this? The honest answers aren't uniformly cynical. Some postings stay live to build a pipeline for a role the company expects to need later. Some survive a hiring freeze that nobody bothered to take down. Some are attached to internal promotions that were decided before the posting ever went up. None of that makes the experience on the other end less frustrating — a candidate spending an evening tailoring a cover letter has no way to tell "this is a real search" from "this is a candidate pool we're quietly building" from "this role was filled internally three weeks ago." The posting looks identical in all three cases.

What employers are seeing

Here's the half of the story that gets less airtime. iHire's 2026 State of Online Recruiting Report — a survey of 1,054 job seekers and 318 employers across 57 industries, released September 22, 2026 — found that 19.2% of employers named fake or fraudulent applicants as their top online-recruiting challenge, and 23.6% said they'd received at least one fake applicant in the past year. "Receiving too many applications from unqualified candidates" was employers' single biggest frustration for the fourth year running, cited by 55.7%.

That volume problem has a measurable, and recent, cause. Greenhouse's own 2026 benchmark study — covering more than 640 million applications across roughly 6,000 companies from 2022 to 2025 — found applications per job had more than doubled, from about 116 in 2022 to roughly 244 in 2025. Ashby, a competing recruiting platform, reports a similar trajectory from a different angle: its customers' recruiters now process around 291 applications per hire, up from roughly 100 in early 2021. Greenhouse's survey work on the candidate side offers a plausible driver — 74% of job seekers say they now personally use AI tools somewhere in their job search, and 49% say they submitted more applications in the past year than the year before. Mass-generate a tailored cover letter in ninety seconds and the marginal cost of applying to one more posting approaches zero. The volume follows.

So: employers facing a tripling of applications per role are, among other things, less able to respond meaningfully to any individual one. And candidates facing silence from roughly 57.5% of the applications they send (iHire's figure for "applied and never heard back" as a top frustration) are, among other things, more inclined to apply faster and wider, trusting fewer individual postings to be worth careful effort. Each side's rational adaptation to the other's behavior makes the other side's complaint more true.

The spiral, not two problems

It's tempting to read ghost jobs and fake applicants as a tit-for-tat morality story — employers started it, or candidates started it, depending who's telling it. The more useful reading is structural. Both ghost postings and AI-assisted mass applications are low-cost ways to keep options open under uncertainty. A company unsure whether it will actually backfill a role can leave the listing up at near-zero cost. A candidate unsure whether any given listing is real can apply to all of them at near-zero cost. Neither behavior requires bad faith. Both behaviors, once common, make the other side's caution rational, which makes the behavior more common. That's a spiral, and spirals don't resolve by exhorting one side to go first.

It also explains why neither side's current self-protective instinct — more filtering on the employer side, more volume on the candidate side — makes the underlying problem better. iHire found that 42.8% of job seekers say it's often hard to tell whether a listing or employer on a platform is legitimate at all. Piling automated rejection on top of that opacity, or piling AI-generated applications on top of recruiter overload, adds noise to a signal that was already thin. Legislatures have started to notice the externality: New York's ban on undisclosed ghost-job postings passed both chambers of the state legislature in 2026, though as of October 13 it had not yet been signed. A law can make the worst version of ghost-posting illegal; it can't by itself make either side trust the process again.

What actually helps — and the limits of that

The data points toward a narrower, less satisfying answer than "ban the bad behavior": the trust gap is, underneath the bad-faith framing, an information problem. Candidates can't see whether a posting corresponds to real intent. Employers can't see whether an application corresponds to a real, qualified person who read the job. What's missing on both sides is a verifiable, structured record of what's actually happening at each stage of a specific application — not a black box that outputs "reject" or "interview" with no visible reasoning, and not a listing that gives no signal about whether anyone behind it is actually deciding.

That's the design argument for hiring pipelines built as genuinely staged, auditable processes rather than a single opaque pass/fail gate. NiceHire's own pipeline is built this way: an applicant clears distinct, independent checkpoints — resume vetting, an AI screening interview, live human interviews — each gating advancement to the next in the schema and service layer, rather than one black-box AI call deciding everything at once. Every candidate is scored against the same fixed criteria — technical, communication, and cultural fit, each out of 100, against a configurable passing threshold — rather than an unexplained score nobody can inspect. And because AI screening interviews run whenever a candidate is available rather than requiring a scheduled slot, the step most prone to silent delay has fewer places to stall.

None of that touches the ghost-job half of the problem directly — a structured pipeline doesn't stop a company from leaving a dead listing live, and no vendor can audit another company's hiring intent from the outside. It also doesn't solve the volume economics driving mass AI applications; a better-structured funnel still has to decide what to do with 244 applications for one role. What it does narrow is the fake-applicant half of the complaint: a pipeline with real, inspectable checkpoints and consistent criteria gives an employer something more defensible to point to than "the algorithm said no." That's a smaller claim than "we fixed hiring trust." It's also the part of the problem that's actually within a hiring platform's control.

Sources

• ResumeBuilder.com, "Ghost Jobs Report" (released October 14, 2026; n=1,011 U.S. adults, fielded August 27, 2026) — via allwork.space
• iHire, "2026 State of Online Recruiting Report" (released September 22, 2026; n=1,054 job seekers, 318 employers) — via HR Dive and the SHRM vendor directory syndication
• Greenhouse platform data on ghost-job rates and application-volume benchmarks, reported via the Wall Street Journal and corroborated independently — via allwork.space and MetaIntro
• New York ghost-job disclosure legislation status as of October 13, 2026 — via Moneywise