AI for HR Generalists: Where to Start

HR generalists should start with Prompt Writing 101, a free 8-minute lesson. It shows the four parts of a request that gets a usable answer, Role, Context, Task and Format, and has you save it as a template. HR writes the same kinds of messages over and over: policy answers, onboarding emails, open enrollment reminders. A good template is the fastest, safest win.

The path is five free lessons and about 47 minutes. Four use ChatGPT only; the last uses n8n. If your job is mostly hiring, the recruiter path goes deeper on that part. The recruiting and HR playbook covers the rules for the whole team.

Which lesson should an HR generalist take first?

Take Prompt Writing 101 first, because HR prompts carry more risk than most. A vague request gets a vague policy answer, and a vague policy answer creates questions you then have to clean up. Structure is how you keep the answer tied to your actual handbook.

How do you answer employee questions without overpromising?

Use the routine in Reply Faster: Customer Emails. You paste the thread with the facts you can stand behind, get a draft, then have ChatGPT list every promise in the draft next to the fact it rests on. It takes 9 minutes. The example is a bakery's customer, but for HR the facts are your policy text, and the check catches a draft that promises time off, a benefit or a date nobody approved.

What does the scorecard lesson teach HR about fairness?

Interview Debrief to Scorecard turns a panel's debrief transcript into a scorecard against the role's written must-haves, with a quote as evidence for each score. In the transcript, a panelist mentions a candidate's young kids, and the scorecard leaves it out, as it should. It takes 9 minutes. That habit, job-related points only, applies to performance notes and investigations too.

When do you need the meeting summary lesson?

Take it for team meetings, benefits planning or onboarding check-ins. Summarize a Meeting Transcript returns a summary, decisions, an action items table, open questions and a follow-up email. It takes 9 minutes. Don't use it on confidential employee-relations meetings unless your company has approved the tool for that.

What does the n8n lesson add?

AI Email Responder builds an n8n agent that saves a reply draft under each new Gmail message, so routine questions start from a draft. Nothing is sent automatically. It takes 12 minutes. Keep it to logistics like where to find the handbook or when payroll runs, never anything about an individual employee.

Which prompts can you copy today?

Three prompts to paste into your company's approved AI tool. Fill in the brackets.

You are an HR generalist at [COMPANY]. Here is our policy text: [PASTE POLICY]. An employee asked: [QUESTION]. Draft a short, friendly answer using only the policy text, and quote the line it comes from. If the policy doesn't answer the question, say so and name who to ask.

Draft an open enrollment reminder for all staff. Enrollment runs [DATES]. What's changing: [CHANGES]. Where to enroll: [LINK]. Who to contact: [CONTACT]. Under 150 words, plain language, no benefits advice.

Turn these onboarding notes into a first-week checklist for a new [ROLE]: [NOTES]. Group by day, name an owner for each item, and mark anything missing an owner as "OWNER NEEDED".

Fuller versions with filled-in examples: employee policy question reply, benefits FAQ for employees, open enrollment reminder email, new hire onboarding checklist and AI use policy draft.

Terms to know: employee onboarding, employee handbook, performance review, AI bias, AI acceptable use policy and PII.

How long will this take, and what do you need?

About 47 minutes in total. The practice versions are free. To work for real:

• Lessons 1 to 4 need ChatGPT, which has free and paid plans. Use a business plan that doesn't train on your data for anything with employee details.
• Lesson 5 needs n8n (n8n Cloud has paid plans, or self-host) and an OpenAI API key, billed separately from ChatGPT plans.

Check each vendor's pricing page for current plans.

What should HR never hand to AI?

Never hand AI an employment decision or an employee's private information.

• No decisions by AI. Hiring, discipline, pay and termination decisions stay with people who can explain them in job-related terms.
• No protected characteristics. Age, family status, health, religion and similar details stay out of prompts and notes.
• Watch for bias. AI can repeat bias from its training data or from your notes. Use written criteria, the same for everyone.
• Get the rules right. This is not legal advice. Follow your company's policies and the law where you operate, and have an employment attorney review how AI fits your process.

Key takeaways

• HR generalists should start with Prompt Writing 101 and save templates for the messages they send most.
• The five-lesson path takes about 47 minutes; four lessons need only ChatGPT.
• The promise check keeps employee replies tied to actual policy text.
• The scorecard lesson's rule, job-related points only, applies across HR work.
• People make employment decisions; keep employee data in approved tools only.

Frequently asked questions

Can I use ChatGPT to write an employee handbook?

It can help you draft and reorganize sections from your existing policies, but have an employment attorney review the result before you publish it.

Is it safe to paste employee information into AI?

Only into a tool your company has approved, on a plan with data controls. Remove names and details the task doesn't need.

What's the difference between this path and the recruiter path?

The recruiter path is about the hiring cycle. This one covers the wider HR job: policy answers, onboarding, benefits messages and meeting notes.