AI made recruiters faster — and left them less sure they hired the right person
The News
AI made finding candidates faster for 72% of US recruiters over the past 12 months, according to TestGorilla's Recruiting in the AI Era report, a survey of 1,428 US talent professionals fielded in August 2026. The speed has not translated into confidence: among recruiters who say AI made them faster, a third reported no gain in confidence that they had found the right person, and 18% said their confidence in the candidate-role match actually fell. Recruiters are filling the gap manually — 67% now spend an hour or more checking a single candidate before trusting they can do the job, and 29% spend two hours or more. Verifying whether claimed skills are real was the single most-cited sourcing challenge, named by 57% of respondents.
Why It Matters
This is the uncomfortable arithmetic of AI sourcing: the tools have shortened one part of the funnel while quietly lengthening another. If your recruiters save twenty minutes on search and then spend an hour verifying what the search returned, your cost-per-hire maths has probably worsened, not improved — and your requisition load certainly hasn't. The report also catches a belief gap: 63% of recruiters think most of the candidate data in their sourcing tools has been independently tested, yet many of that same group report claimed skills turning out weaker than advertised. Treat this as a due-diligence prompt on your own stack before it becomes a quality-of-hire problem.
Taking Action
All action is situated in your corporate context. These are starting points, not prescriptions — choose what fits.
Connect
Who to talk with in your enterprise
Your director of recruiting: ask how long recruiters actually spend verifying AI-sourced candidates against self-reported profiles — the survey suggests an hour or more per candidate is now normal. Your head of talent analytics: compare time-to-fill trends with quality-of-hire metrics (90-day performance, early attrition) to check whether the speed gain is real or simply moved downstream.
Inspect
What to take a closer look at
Audit what your sourcing tools claim is 'verified' data. Pull a sample of twenty recent hires sourced through AI tools and check claimed skills against assessment results or manager feedback at 90 days. If skills are holding up worse than advertised for your recruiters too, pilot an objective pre-interview check — 92% of surveyed recruiters said they would run one, and only 4% would go straight to interview.
Reflect
What to think about more deeply
If your AI sourcing tool needs an hour of human verification per candidate to be trusted, is it a hiring tool or a hypothesis generator?