The headline: AI adoption has crossed the tipping point
For the first time in our annual survey, more than two-thirds of HR professionals report using some form of AI in their hiring or people operations workflows. That's up from 41% in 2024 and represents the fastest adoption curve we've seen for any HR technology category.
But adoption doesn't mean satisfaction. The story beneath the headline numbers is more nuanced - and more useful.
Using AI in hiring
Up from 41% in 2024. The majority have moved from "exploring" to "actively using."
Satisfied with their AI tools
Less than half of users say their current tools are meeting expectations. There's a quality gap.
Plan to increase AI spend
Despite mixed satisfaction, most teams are investing more. They're looking for better tools, not fewer.
Average ROI on AI screening
Teams using AI screening report 3.2x return vs. manual process - measured in recruiter time saved.
What tools are actually being used
What's working and what isn't
The winners
- AI CV screening: The category with the highest satisfaction score (71%). The value proposition - volume reduction with quality maintenance - is clear and measurable.
- JD writing assistance: High adoption (61%) and solid satisfaction (65%). Significantly reduces time on a task most recruiters find tedious.
- Onboarding automation: Lower adoption but among the highest satisfaction scores (74%) of any category. Teams that have implemented it rarely go back.
The underperformers
- AI interview scoring: The most polarising category. 34% of users report it as "not meeting expectations." Quality varies enormously by vendor.
- Predictive attrition: High interest, low satisfaction. The data requirements are more demanding than most teams have available.
- "AI-native" ATS platforms: Many teams adopted new ATS platforms on the promise of AI features - and found the AI was marketing, not substance.
"We bought an AI-native ATS for the screening features. Eighteen months in, we're still doing most screening manually because the AI recommendations aren't good enough to trust." - Head of TA, Series C SaaS company
The bias question
We asked directly: has your organisation assessed your AI hiring tools for bias? The results were sobering.
- Only 31% had conducted any form of bias audit on their AI hiring tools
- 54% said they assumed their vendor had handled this - without verifying
- 15% said they hadn't thought about it
This is an area of significant risk. Several jurisdictions (including the EU AI Act and New York City Local Law 144) now require documented bias assessments for AI tools used in employment decisions. Ignorance is not a defence.
What high-performing teams do differently
When we looked at the teams reporting the highest satisfaction and ROI from their AI tools, a few patterns emerged consistently:
- They treat AI criteria as a living document - reviewing and updating after every batch of hires
- They use AI scores as a ranked starting point, not a binary pass/fail cutoff
- They measure 12-month retention for AI-screened hires and use it to calibrate the model
- They've told candidates that AI is used in screening (and report no negative impact on application rates)
- They have a named internal owner for AI hiring tool performance - not "the recruiter who set it up"
See how PeoplePanda performs against this benchmark
Start a free trial and get your first AI-screened shortlist within 24 hours.
Get started free →