AI for HR Teams: Where It Helps, Where It Is a Legal Landmine
A careful guide to using AI in HR — hiring, feedback, policy — with clear lines on what should never touch a model.
HR is one of the most sensitive places to introduce AI. It is also one of the highest-leverage. The rule of thumb: use AI on process, never on people. Anything that touches a decision about a specific human — hire, fire, promote, discipline — needs a human accountable and a paper trail. Everything else is fair game.
Where AI clearly helps
- Rewriting policies in plain language so employees actually read them.
- Drafting job descriptions from a hiring manager's messy notes.
- Summarising engagement survey comments into themes without exposing individuals.
- Turning legal updates into short internal briefings the whole company can understand.
Where AI is a legal landmine
Do not use AI to score candidates, rank employees, or make disciplinary decisions without careful, audited human oversight. In many jurisdictions this is either regulated or outright illegal, and even where it is legal it is a lawsuit waiting to happen. A model that filters resumes on patterns you cannot inspect is a discrimination case with a friendly interface.
A safe workflow for hiring
- AI drafts job posts and interview questions from your criteria.
- Humans screen resumes and conduct interviews.
- AI helps summarise notes from multiple interviewers into a single view.
- Humans make the decision and document their reasoning without AI phrasing it for them.
Data hygiene for HR
- Never paste employee names, IDs, salaries, or medical info into a public model.
- Use a private, approved AI workspace with audit logs.
- Keep human-written notes for every people decision. If it is ever contested, AI-generated reasoning will not hold up.
In HR, the goal of AI is to give your team more hours for the human conversations, not fewer.