Editorial note: Examples on this page are illustrative unless a company, study or source is specifically named. The aim is to show realistic workflows without presenting invented scenarios as case studies.
Where AI is genuinely useful right now
- Résumé screening for volume roles. Sorting large applicant pools by basic qualification criteria before human review.
- Interview scheduling. Coordinating calendars across candidates and interview panels automatically.
- Onboarding support. Answering new-hire questions about policies and processes on demand.
- Job description drafting. A faster first draft, refined for tone and to avoid exclusionary language.
Illustrative scenario: large employers use AI screening tools to sort thousands of applications for high-volume roles down to a manageable shortlist, with recruiters making all final decisions.
Where to be careful
- Discrimination risk in screening. AI hiring tools have been shown to inherit biases from historical hiring data — several US states and the EU now specifically regulate AI use in employment decisions.
- Never let AI make the final hiring decision. Screening support is different from decision-making — a human should always make the actual call.
- Candidate disclosure. Some jurisdictions require telling candidates when AI is used to evaluate them — check local requirements (e.g. NYC's AI hiring law).
Worth doing: periodically audit your AI screening tool's outcomes by demographic group, even informally — patterns that look neutral in the tool's design can still produce skewed results in practice.
Getting started: If you're introducing AI into hiring, start with logistics like scheduling and answering candidate FAQs before touching anything that filters or ranks candidates, since that's where regulatory and bias risk concentrates.
A practical starting workflow for HR teams
- Start with resume screening support, not final decisions. Use AI to surface candidates for human review rather than to auto-reject applicants — this reduces both bias risk and legal exposure.
- Audit for bias before wide rollout. Run a test batch of past hiring decisions through the tool and check whether it disproportionately screens out any group compared to your actual human decisions.
- Check hiring-AI regulations for your region. Several jurisdictions now require disclosure or audits when AI is used in hiring decisions — rules vary and are still developing.
- Keep a human in every final decision. Beyond legal risk, candidates generally expect a human made the actual call on their employment.
Let AI prepare HR work, not make the people decision
HR teams can save time drafting job descriptions, grouping survey comments and preparing interview questions. I would not use a general AI model as an unsupervised gatekeeper for hiring, promotion or disciplinary decisions because bias and missing context are difficult to see in a polished score.
A practical test: A safer pilot is to use AI after a human has defined the criteria: generate structured interview questions from those criteria, then have HR review them before use.
A prompt template to adapt
This is a template, not a tested result. Replace the brackets with your own details:
"Draft a clear, neutral job description for a [role] based on these duties: [duties]. Use inclusive language, list only requirements that are genuinely needed, and suggest five structured interview questions tied to those duties."
What to check before you rely on it
- Hiring, promotion and discipline are high-stakes decisions. Keep a person responsible, and be cautious about any tool that scores or ranks people.
- Check the law where you operate. Rules on automated decisions in employment are tightening in several regions (see The Global AI Regulation Landscape).
- Keep employee personal details, performance reviews and health information out of general AI tools.
- Test outputs for bias, for example by comparing results for similar profiles with different names.
For the habits behind these checks, see Fact-Checking AI Output and Privacy & Data When Using AI.
Frequently asked questions
Do candidates have to be told AI is used in hiring?
Increasingly, yes, in specific jurisdictions — New York City's law is a well-known example. Requirements vary by location, so check local rules before deploying AI screening.
Can AI eliminate bias in hiring?
Not automatically — AI models trained on historical hiring data can actually reproduce or amplify past biases. Genuine bias reduction requires deliberate model auditing, not just automation.