Recruiters are drowning and AI isn't saving them yet

Artificial intelligence was supposed to make hiring easier. For most recruiters, it has made it harder.

Despite a flood of AI-powered tools entering the market, a growing body of research and first-hand industry experience suggests the technology is not yet delivering the productivity gains businesses were promised and in many cases, it is actively slowing teams down.

That is the sobering assessment shared by Jamie Kohn, senior director of research in Gartner's HR Practice, who says the data tells a clear story.

"AI is not having a significant impact on recruiter productivity yet," she said. "We have added tools, but we have not redefined roles and workflows. We're just adding more steps to the process."

Tom Hyde, head of sales and revenue APAC at recruiting software company Greenhouse, echoes the sentiment.

After processing data from more than 7,000 businesses and 22 million monthly job applications, Greenhouse has arrived at a troubling conclusion: for the first time in its 14-year history, both candidates and employers are unhappy with the hiring process at the same time.

"It's creating what we call an AI doom loop," Hyde said. "Candidates are using AI to mass apply for roles. Businesses are using AI to sift through the massive volume. And good signals are getting lost on both sides."

The productivity promise that hasn't arrived

The scale of the problem is becoming stark. Hyde said average recruiter application volumes are up more than 400% year on year – a direct consequence of AI tools enabling candidates to apply to dozens of roles in minutes. Yet the tools designed to help recruiters manage that volume are not yet fit for purpose.

Kohn detailed an example from a talent acquisition leader who tried to introduce AI-assisted job descriptions to her team.

"They would go to the HCM, copy the job description, take it over to the gen AI, paste it in with a prompt, then copy and paste it back," Kohn said. "There was a lot of manual work involved. It's not saving them time – it's clunky."

For Kohn, the fix is structural. "For us to get productivity gains, recruiters will have to be removed from steps of the process – not removed from decisions, but removed from having to go into the software and put a prompt in."

She pointed to emerging job-posting agents that can automatically generate a draft from hiring requirements without requiring recruiter input at every step. That kind of embedded AI, she argues, is what will genuinely shift the needle.

The distinction between high-volume and high-complexity hiring is also critical. Kohn said frontline, high-volume roles like retail, customer service, warehousing, and drivers are where agentic AI has the clearest near-term case.

"Your screening criteria are very straightforward. It's very easy to hand that off to AI." For competitive or niche roles, however, the benefit is less about capacity and more about quality.

Structured hiring as the antidote

For Hyde, the businesses getting AI right are not those bolting it onto existing processes, they are those who have first built a strong hiring foundation.

"The businesses that are doing it really well are overlaying AI on top of already solid hiring principles and good structure," he said. "Those that don't have that process nailed down are trying to use AI to combat large volume, and that's where it's falling down."

Greenhouse's own research shows the business case for getting this right is compelling. Companies that bring consistency and structure to their interview process see a 53% decrease in the need to rehire within the first 12 months.

With Hyde noting the tangible cost of a failed hire is close to $280,000 – factoring in lost productivity, morale, and onboarding – that is a significant saving.

Nine out of 10 candidates surveyed by Greenhouse said they expect to get a genuine feel for a company's culture and values during the interview process – yet many come away disappointed.

Hyde said businesses that are proactively using AI to personalise follow-ups, provide transparent communication, and create consistent interview experiences are separating themselves from the pack.

"Candidates aren't anti-AI," Hyde stressed. "A very small percentage in the survey said they wanted to see less of it. They just want to see it used in a better way – with transparency upfront about how it's being used and what it's measuring."

When AI fights AI, everyone loses

Perhaps the most striking dynamic to emerge from both conversations is what Kohn calls a troubling stand-off between AI-assisted candidates and AI-screening employers.

"If everyone puts their CV through generative AI, then everybody's CV looks the same," she said. "We cannot continue down this path of fighting AI with AI forever."

Kohn has heard some recruiters argue the inbound applications pipeline is now so compromised that proactive sourcing is the only reliable channel. Others are experimenting with AI interview agents to conduct brief initial conversations with all applicants before any human review. But she is candid about where the industry stands: "Right now, recruiters are drowning. And while they're drowning, they're being told, 'Just use AI, that'll be fine.'"

The challenge is particularly worrying for younger jobseekers. Greenhouse data revealed that Gen Z candidates are disproportionately affected by the current dysfunction – not because they lack digital fluency, but because they lack professional networks.

"If you don't have that referral network to help open doors, and you're looking to break into the job market in your first or second role, you've got to work out how to make your application stand out in a sea of noise," said Hyde.

AI bias also remains a live concern. Kohn noted that many predictive hiring models are trained on historical hire data – an inherently limited dataset that excludes everyone who was passed over but could have succeeded.

She does, however, see cause for optimism in AI-enabled skills inference tools that can look beyond stated qualifications to infer adjacent capabilities. "Instead of ruling you out, it's ruling you in.”

What TA leaders should do now

Both Kohn and Hyde offered practical guidance for HR and talent acquisition leaders navigating the current environment. Kohn's first piece of advice is blunt: stop overselling AI internally.

"Leaders are not realistic about what AI can and can't do right now. Don't try to make the business case for AI based on productivity – it's not there yet. You will end up overstretching your recruiters,” explained Kohn.

Her second recommendation is to split expectations by hiring type, with a faster AI-forward push for high-volume roles and a more measured, augmented approach for complex ones.

Third – and most importantly – she urged leaders to redesign workflows with recruiters, not just for them. "Recruiters are brilliant at identifying where workflows can be changed. That will help them free up more time to evaluate candidates and build relationships."

Hyde's message to businesses is similarly grounded. "Focus on the intake – how you're treating candidates – because that will have a superb effect down the track. The good news is it's very fixable. We've worked with a lot of organisations that have turned around quite an antiquated, lumpy hiring process into one where they're actually attracting great talent and really happy with the people they're getting through the door."

For Kohn, the longer-term picture is genuinely promising, but contingent on a fundamental rethink of how talent acquisition operates. She pointed to Gartner data showing that while AI has had no measurable effect on recruiter productivity, it has had a positive effect on quality of hire and consultative recruiter behaviour.

Recruiters who used AI more frequently were more likely to recommend a different sourcing strategy to hiring managers, suggest changes to hiring criteria, and help rewrite job descriptions.

That, she argues, is the real signal and the direction in which the profession is heading. "Where success factors are not straightforward, we can use technology to get a better handle on what our hiring criteria should be, to expand the talent pool. Recruiters become even more important in the future – not less."