What are AI agents for HR? (And where they break)

What an HR agent actually does

An AI agent for HR takes an employee request all the way to done. Someone asks how much leave they have left, or needs an employment letter for a landlord. The agent reads the request, checks it against the policy your team already wrote down, makes the change or produces the document in your HR system, and hands the case to a person when the policy does not cover it. The HR chatbot most teams already run stops one step earlier: it quotes the policy and leaves the filing to whoever asked.

That one step is where most rollouts stall. An agent that can read your HR system but not write to it still needs a person to press the button, adding a handoff instead of removing one.

The published numbers are not the numbers you get

Vendors publish resolution rates between 40% and 75%. Measured the strict way, where no person touched the case and the employee did not come back within 72 hours, the industry average is 20% to 30%, and 40% to 60% is best in class after a year or two of tuning. Plan the business case on the lower figure. Supervision is real work too: one 2026 survey put it at over 6 hours a week per person spent checking AI output.

Start with service requests, not hiring

Pick one queue where the policy is already written: leave balances, letter requests, onboarding paperwork. Put hiring last. New York City requires a bias audit less than a year old plus 10 business days of notice to candidates, at $500 to $1,500 per violation per day, and the EU deadline for hiring tools moved from 2 August 2026 to 2 December 2027.