Gartner: AI Agents May Represent 1 in 4 Candidates by 2029
Your daily brief on leadership and business
By 2029, one in four candidates may be represented by a machine, according to Gartner research. That means an agent acting for the candidate, not a tool the candidate uses. Jamie Kohn, a senior director analyst in Gartner's HR practice, said so on Monday at the Gartner HR Symposium/Xpo.
Talent acquisition teams tend to lead HR in deploying AI tools, Kohn said. They also face contrasting demands from employers and candidates. AI, risk management and hiring for skills will collide, she said, and shape the talent strategy. She laid out four trends.
Machines as an audience
More people now lean on generative AI to make decisions, including about jobs. "Machines increasingly determine whether candidates hear about us at all," Kohn said. A company may receive applications from people who have never heard of its job or brand, because AI made that call from information the company does not fully control. The tools draw on news coverage, Reddit and employee reviews on third-party sites, not only the career page.
Her advice: publish clear, structured, text-based content, such as videos with readable transcripts. Keep it current on topics like flexibility. Be specific, back claims with data and employee stories, and answer candidates' questions in full. That is a shift from older demands to keep job information short.
From avoiding risk to being aware of it
Kohn said employers used to try to avoid AI risk in candidate assessment. Now more companies use AI to scale their work and accept the risk that comes with it, partly because refusing could hamstring a team. Application volume has more than doubled in the last two years, she said, and data from other companies points to a near tripling, due in part to bots and AI agents. Organizations also say they are working harder to find high-quality candidates.
The "firewall" around screening and assessment is crumbling, she said. Recruiters will use AI "whether we approve it or not," and keeping it out of screening forever is "just not sustainable." Roles that are stable and repeatable, with clear criteria and large scale, may be lower-risk places to speed up, for example through AI interviews. Risk rises when skills are hard to observe or success is ambiguous.
Building their own tools
Recruiters no longer have to push a vendor through procurement or ask IT for a new solution. They get new features from current vendors and from enterprise AI tools, and more of them want to build their own. "Every year brings new tech capabilities that are accessible almost instantly," Kohn said. HR tech leaders reported building AI agents directly in their human capital management platforms, often with no-code and low-code tools. That can lead to several tools doing the same job, uneven processes and technical debt. "The challenge is not a lack of innovation. It's keeping up with it," she said.
Employers should weigh how widely a tool will be used and how risky that use is. Labor market analysis is an easy one for recruiters to craft themselves. Candidate matching and rediscovery may need more oversight. Assessments should be managed at the enterprise level because of their reach and risk, per the presentation.
New yardsticks for recruiters
Recruiters are dealing with talent shortages and trouble attracting talent. They are also pushed to move fast, defer to hiring managers and focus only on open roles. "We're asking recruiters to behave strategically while surrounding them with a system that enables and rewards the wrong behavior," Kohn said.
Her fix is to rethink process design, data access and partnerships, and to measure what employers actually want recruiters to do. She named slate quality, time to productivity and candidate confidence. "If recruiters have to fight the system to be strategic, the system's probably going to win," she said.
Source: HR Dive