HR sits at the heart of every organisation — hiring, welcoming new employees, handling sensitive conversations, writing policies, and keeping people informed. It is also a role buried in writing tasks: job postings, offer letters, onboarding checklists, policy updates, performance review language, and internal announcements.
That is exactly where AI assistants shine. Tools like ChatGPT, Claude, and Gemini are not here to make HR decisions for you. They are more like a capable assistant who can draft the first version of a document while you focus on the judgment calls that only a human can make. Here is a grounded look at how people in HR commonly use these tools today.
Before you start: never paste real employee names, salary figures, medical information, or other personal details into a public AI tool. Use general descriptions instead — "a mid-level software engineer" rather than a person's name. Your company may also have its own data policy on AI use, so check before you begin.
What's in this guide
- 1. Writing and improving job postings
- 2. Drafting interview questions
- 3. Building onboarding materials
- 4. Drafting and refreshing HR policies
- 5. Performance review language
- 6. Internal communications and announcements
- 7. Employee FAQ documents
- 8. Researching HR topics and best practices
- Common worries, answered
- Frequently Asked Questions
1. Writing and improving job postings
Describe the role, the main responsibilities, and the kind of person you are looking for, and an AI assistant will produce a solid first draft in seconds. You can also paste in an existing posting and ask it to "make this more inclusive" or "make this shorter and punchier."
Example prompt: "Write a job posting for a part-time customer service coordinator. The person will handle phone and email enquiries, update our CRM, and escalate complaints to the manager. Friendly, professional tone."
Caution: AI can sometimes include phrasing that unintentionally favours certain groups. Always read the final posting carefully and consider running it through a bias-checking tool or having a second person review it before publishing.
2. Drafting interview questions
Building a consistent set of interview questions for every candidate is good practice and often a legal protection. AI can generate a list of behavioural questions, situational questions, or skills-based questions for almost any role you describe.
Example prompt: "Give me eight behavioural interview questions for a project manager role focused on communication, deadline management, and handling conflict."
Caution: Some interview questions are legally off-limits in many countries (for example, questions about age, family status, or health). AI does not always know your local employment law. Always have a qualified person review interview questions before use, and consult your legal team or an employment law resource for guidance on what is and is not permitted.
3. Building onboarding materials
AI can help you build first-day checklists, welcome emails, "who to contact for what" guides, and simple orientation schedules. Give it your company's structure in general terms and it will fill in the blanks you can then customise.
Example prompt: "Write a friendly welcome email for a new employee starting on Monday. Mention that their manager will meet them at reception, where to find the IT setup guide, and that lunch is informal on the first day."
Caution: Onboarding documents often reference specific systems, benefit providers, and contact names. AI will not know these details — you will need to fill them in. Sending an onboarding email with placeholder text is easy to do and hard to walk back.
4. Drafting and refreshing HR policies
Starting a new policy from a blank page is daunting. AI can produce a structured draft of common HR policies — remote work guidelines, expenses policies, code of conduct sections — that you then adapt to your organisation's specifics and get reviewed by the right people.
Example prompt: "Draft a remote work policy for a small company. Cover eligibility, expected availability hours, equipment responsibilities, and data security basics."
Caution: HR policies carry legal weight. An AI draft is a starting point, not a finished document. Always have a qualified employment lawyer or HR specialist review any policy before it becomes official.
5. Performance review language
Finding the right words for a performance review — especially a difficult one — is genuinely hard. AI can suggest neutral, constructive phrases when you describe a situation in general terms, helping you avoid language that is either too vague or accidentally dismissive.
Example prompt: "Suggest three ways to phrase feedback for someone who consistently meets deadlines but struggles to communicate progress to the rest of the team. Keep the tone constructive."
Caution: Never include the employee's name or identifying details in a prompt to a public AI tool. Use the AI's suggestion as a starting point and always personalise it — a review that reads like it was written by a machine is obvious and can feel dismissive.
6. Internal communications and announcements
Announcing a policy change, a new benefit, or a company update requires a tone that is clear, warm, and consistent. AI can draft announcements quickly, and you can ask it to adjust the reading level or formality to suit different audiences.
Example prompt: "Write a short all-staff email announcing that the company is moving to a four-and-a-half-day work week starting next month. Friendly and positive tone, under 150 words."
Caution: Internal communications reflect your organisation's culture and leadership voice. Have someone senior review anything that touches sensitive topics — redundancies, restructures, or benefit changes — before it goes out.
7. Employee FAQ documents
When a policy changes or a new process launches, employees always have the same questions. AI can turn a policy document or a set of bullet points into a friendly FAQ page that is much easier for people to read and actually use.
Example prompt: "Turn these five bullet points about our new expenses process into a friendly FAQ with plain-language answers. [paste bullet points]"
Caution: Check every answer the AI produces against the actual policy. It will do its best to interpret your bullet points, but may fill gaps with assumptions. Any FAQ that goes to employees should be verified line by line.
8. Researching HR topics and best practices
Need to get up to speed on a topic — flexible working trends, approaches to managing neurodiversity in the workplace, or how other companies handle hybrid office scheduling? AI can give you a useful overview and point you toward the kinds of questions worth exploring further.
Example prompt: "Summarise the main approaches companies use to support neurodivergent employees at work. Give me a practical overview I can use as a starting point for a team discussion."
Caution: AI can produce plausible-sounding information that is outdated or simply wrong. Treat AI research summaries as a starting point, not a final source. Verify key facts through professional HR bodies, government guidance, or a subject-matter expert before acting on them.
Common worries, answered
Many HR professionals worry that using AI means giving up control — or that it signals to their employer that the job could be done by a machine. Neither is true. AI handles the blank-page problem and the repetitive writing load. It does not handle the conversation with a grieving employee, the judgment call on a disciplinary matter, or the relationship-building that keeps people from leaving. Those things are irreplaceable, and they are the core of what HR professionals do. Using AI for the admin side means you have more time and energy for the parts that actually require you.
The other common worry is data privacy — and that one is worth taking seriously. The rule is simple: describe situations in general terms, never paste in personal information, and check your organisation's policy on AI tool use. When you follow those boundaries, AI is a genuinely useful and safe time-saver.
Frequently Asked Questions
Is it safe to use AI for HR tasks?
AI can be a helpful assistant for drafting, summarising, and brainstorming, but you should never paste employee names, salaries, or personal details into a public AI tool. Always verify any output before using it officially, and keep a human making the final decisions.
Can AI write job postings for me?
Yes. Most major AI assistants can draft a job posting from a simple description of the role. You will still need to review it for accuracy, adjust the tone to match your company, and check for any unintentional bias in the language before publishing.
Will AI replace HR professionals?
No. HR work depends on human judgment, empathy, legal knowledge, and trust — things AI cannot replicate. AI handles the repetitive writing and research tasks so you have more time for the human side of your work.
Can AI help with performance review language?
Yes. AI is good at suggesting neutral, constructive phrases when you describe a situation in general terms. Just avoid including the employee's real name or identifying details, and always personalise the output before using it.
What AI tools do HR professionals commonly use?
Many HR professionals use general-purpose AI assistants like ChatGPT, Claude, or Gemini for writing and research tasks. Some HR software platforms are also beginning to include built-in AI features for things like candidate screening summaries and policy drafts.
Not sure where to start with AI?
Take our two-minute quiz and get a personalised recommendation for your situation — no jargon, no pressure.
Take the Quiz Browse All Guides