This could be the year AI automation takes over HR

Reasons and evidence for an AI HR transformation

Besides consulting and market analysis, Bersin's company has been in the AI business since late 2023, when it introduced Galileo, a GenAI assistant for HR practitioners, and added agentic AI features in 2025. It recently developed a model that identifies 94 typical capabilities of HR workers and mapped them to around 120 possible agents. It plans this year to stitch together the agents into around 10 "super agents" that can handle common tasks autonomously from start to finish.

Bersin uses a car metaphor to illustrate the potential leap in HR automation. A ChatGPT app like Microsoft Copilot is a personal productivity tool analogous to power steering or brakes. It's nowhere near to being a driverless car.

Holger Mueller, vice president and principal analyst at Constellation Research, says HR isn't just fertile ground for AI-driven automation; it's the vanguard for the entire organization.

Holger Mueller

"They're going to be the showcase, the first mover, the place where AI will be adopted in the enterprise," Mueller said.

Other departments currently undergoing automation through AI, such as manufacturing and finance, are subsets of the organization, whereas HR processes touch every employee, Mueller said.

"If I make every manufacturing employee 20% more productive, that's great. But that's only one-third of my people. If I can make everybody 5% more productive, I move the yardstick much more for the enterprise," he said.

Focusing on automating common administrative tasks, such as changing employee direct-deposit information or preparing a job requisition, can have a big effect. "Those are things that happen tens, hundreds, thousands of times," Mueller said. "If I squeeze 50% out of this, 30% out of that, it's massive scale." He predicted software vendors will introduce conversational UIs this year that can automate certain HR tasks.

But Chris Hester, senior director analyst at Gartner, is less sanguine about a major HR transformation happening this year, especially an agentic AI-driven one. Gartner client inquiries suggest midsize and large enterprises are only in the early stages of the journey.

Chris Hester

"The majority of our clients are still very much in a place where they are sort of experimenting with AI, and really with generative AI," Hester said, and many are starting pilots.

Gartner surveys also indicate mixed results with AI. "Our data is showing that while 64% of HR leaders who have invested in AI have reported a positive impact on HR productivity over the past year, only 1% strongly agree and 24% somewhat agree that AI investments have reduced overall HR costs," Hester said. "To me, this indicates that productivity is potentially being misapplied to low-value tasks."

He nevertheless agrees HR departments will start to see productivity gains and cost efficiencies from agentic AI, noting 89% of HR leaders said they plan to pursue some form of agentic AI in 2026.

Agent integration will be the biggest hurdle

Bersin cautioned that companies will face integration hurdles because many of the 94 HR capabilities are disconnected. "Ideally, they should be operating in an integrated system, so there's going to be a ton of re-engineering going on in HR," he said.

Interoperability will improve by the end of the year but remains a big unknown, including for HR software vendors, Bersin said. "They're all presuming that MCP [Model Context Protocol] will solve their interoperability problems, but I haven't seen it yet."

Hester pointed out that some organizations adopting agentic AI have used MCP, which enables AI applications to communicate with external services. But MCP carries some risk, can be costly and requires maintaining unique connections, he said. And while some enterprise software vendors offer agentic AI tools that enable multi-agent systems, the ability of agents to work across platforms is limited. The same goes for working across departments like HR and finance.

MCP has enabled some of this work, as has Google's agent-to-agent (A2A) protocol that several vendors signed onto, but few have retooled their agents and architectures to follow the standard, Hester said.

If major AI players like AWS, Google and Microsoft can agree on and implement an A2A communication standard, application vendors will likely follow, making enterprise-wide agentic AI feasible. "That is when I think we will truly start to see significant advancements and shifts in major workflows and business processes," he said.

4 HR automation steps business and IT leaders can take now

These experts offered advice on how to prepare for AI transformation in HR.

1. Formulate a strategy

"The majority of CHROs understand that they need to lead and champion AI within the HR organization," Hester said. But while a slight majority of Gartner clients said they have piloted projects, only 20% have a strategy.

According to Hester, to champion HR automation effectively, CHROs should do the following:

• Have a vision for how AI could transform HR.
• Give senior leaders enough training to understand AI fundamentals. Without basic knowledge, it's hard to come up with a vision and brainstorm applications that are tied to business value and objectives.
• Coordinate with other departments.
• Think about long-term transformation, not just short-term, everyday capabilities. Consider how AI could change how work gets done. "Think about the workflows," Hester said. "Can you simplify them? Can they be automated in some way, shape or form through AI? Where are the biggest workflow opportunities?"
• Decide what HR's role will be in companywide training initiatives and in advising business leaders on how the workforce could be transformed with AI.

2. Educate the C-suite about AI's potential

Hester said recent news tying layoffs to AI had led boards and CEOs to focus too much on the potential to cut labor costs. "Actually, less than 1% of layoffs in the first half of 2025 resulted from AI increasing or modifying workflows or employee productivity," he said. "Full, massive restructuring of the enterprise, organizational design that truly aligns with AI, or significant cost efficiencies have yet to really be realized by the majority of organizations."

3. Identify needed AI skills

To make the leap to more comprehensive automation, business leaders must be aware of the AI skills they need, then assemble teams that understand HR processes well enough to build super agents to handle each process, Bersin said.

While some programmers and software engineers are needed, much of the work can be done by subject matter experts using natural language programming tools.

"I think 80%, maybe 90% of the new skills that are needed are understanding the general operating model of how AI works, which most of us are learning on our own by using ChatGPT," Bersin said. He likened the skill level to programming Microsoft Excel macros.

Much of the work will be in retraining AI by feeding it corporate documents so it can answer questions accurately and in learning how to craft AI prompts that produce useful responses.

"Every company's processes are a little bit different, so even with an off-the-shelf agent that might seem like it's perfectly suited to do this, somebody is going to have to get their hands dirty," he said.

Hester recommended building in enough time for employees to gain experience using AI. Sentiment analysis and other engagement methods can guide the upskilling strategy.

4. Reinvent your HR career for the AI future

Bersin said people who do work that is increasingly being taken over by AI -- such as video production, candidate screening and interviewing, and onboarding -- must be willing to reinvent their careers to make AI skills a priority or risk falling behind. It will involve being sufficiently aware of HR processes to consider ways to improve them with AI.

"Whatever thing you do in HR, if you're not thinking about what's the AI way to do that thing, you're going to look around one day and say, 'I'm not getting a promotion this year. I wonder why?'"

Indeed, all three analysts emphasized gaining a thorough understanding of HR business processes as a prerequisite of any AI automation strategy.

"Companies that really think about the workflows and how work gets done and apply AI in a targeted way to high value business processes across the organization are twice as likely to exceed revenue goals," Hester said. "That is proven research."

David Essex is an industry editor who creates in-depth content on enterprise applications, emerging technology and market trends for several Informa TechTarget websites.