The average job posting on Greenhouse now draws 254 applicants, and the number of applications landing on each recruiter's desk is up 412% in a year, largely because candidates are using AI to mass-apply while employers use AI to mass-filter. Greenhouse's own CEO calls the result a "doom loop." For hiring leaders, the practical effect is that raw application volume has stopped meaning anything, and the processes built around counting it need to change.
What's actually happening in the application "doom loop"?
It's a feedback loop, not a one-time spike. Job seekers use AI tools to generate and tailor resumes for dozens of roles in minutes. Recruiters, facing that volume, lean harder on AI screening to triage it. Each side's tooling escalates in response to the other's, and the net effect is more noise for everyone, not more signal.
Why more applicants doesn't mean more good candidates
Greenhouse now hosts roughly 175,000 live job postings, and nearly 500,000 applications have gone through its AI-assisted "My Dream Job" feature, which matches candidates to roles rather than blasting the same resume everywhere. Candidates who use it get hired at roughly 5 times the rate of other applicants. That's the tell: volume and intent are two different things, and most application-tracking metrics still only measure the first one.
The trust gap runs the other direction too
It isn't only recruiters who are overwhelmed. Among candidates who have gone through an AI-run interview, 63% say they've now faced one, up 13 points in six months, and 38% say they've walked away from a hiring process specifically because of how AI was used in it. Only 18% believe their employer has a clear AI policy, and 57% think disclosure should be legally required, not optional.
When both sides automate around each other instead of talking to each other, application volume goes up and signal goes to zero.
On what breaks when applying and screening both go fully automated
What actually cuts through the noise
The fix isn't a better filter for a bigger pile. It's not needing the pile to tell you anything in the first place. A hiring process built on a defined scorecard and a consistent, structured interview doesn't care whether a role drew 40 applicants or 2,500, because the evaluation happens at the conversation, not the resume screen. That also solves the trust half of the problem: a structured interview is disclosed, explainable, and produces the same kind of evidence a candidate would want to see if they asked how the decision was made.
The practical takeaway: stop trying to build a better filter for the flood, and start building an interview process that doesn't depend on the resume pile being a meaningful signal to begin with.
Where to start if your requisitions feel like they're drowning
- Separate "applied" from "qualified" in your reporting. A rising application count with a flat qualified-candidate count is a screening problem, not a sourcing win.
- Tell candidates plainly where and how AI is used in your process. It costs little and, per the numbers above, is what most candidates say they want anyway.
- Weight structured, scored interviews more heavily than resume-matching for roles competing with high-profile brand names, where application volume is most inflated and least informative.
- For hard-to-fill portfolio company roles, consider targeted or referral-based sourcing over broad job board posting, since less prominent employers see far less AI-generated application noise per opening.
The bottom line for hiring leaders
Application volume in 2026 is no longer a proxy for interest or fit, on either side of the process. The hiring teams staying sane are the ones who've stopped trying to out-filter the flood and instead moved the real evaluation to a structured interview that produces its own evidence, regardless of how many resumes came in the door.