ChatGPT for recruiting: what to hand it and what to keep

ChatGPT for recruiting works best as a drafting assistant for job posts, search strings, outreach, and interview questions, not as a decision maker.

It cuts the blank-page time on tasks you repeat every week, which frees hours for calls with candidates and hiring managers. But it also makes confident errors, so you need a clear line on what it touches and what it never sees.

ChatGPT for recruiting today: where teams use it

Recruiting is where HR teams use AI the most, and the leading uses are job postings, resume screening, and interview scheduling. In its 2026 research, SHRM surveyed 1,908 HR professionals and found 27% of organizations apply AI in talent acquisition, more than any other HR function. Those respondents are HR professionals, not agency recruiters, so agency and independent recruiters should treat the figure as a direction rather than a benchmark.

The same SHRM study lists the most common uses as writing and optimizing job postings, screening and matching resumes, and scheduling interviews. A SHRM researcher said these tasks "relieve a considerable amount of time" for recruiters. You can spend that time on intake calls, candidate conversations, and client updates.

Drafting tasks it handles well

ChatGPT does well on first drafts of text you'd otherwise write from scratch, as long as you supply the facts. These recruiting tasks fit that pattern well.

• Job posts: It turns your intake notes into a structured draft you can edit. Start from your own job description template so the must-haves stay accurate.
• Search strings: It builds boolean search strings, which combine keywords with AND, OR, and NOT, from a plain list of titles and skills. It can also draft X-ray search queries, which use a search engine's site: operator to find profiles on one public website.
• Outreach and follow-ups: It drafts first messages and polite nudges in different tones, which helps when you recruit for a nurse manager one hour and a sales lead the next.
• Interview question banks: It turns a role's core skills into behavioral and technical questions for roles from warehouse supervisor to backend engineer.
• Summaries of your own notes: It condenses intake or debrief notes into a short brief for the hiring manager or client, once you strip out names and identifying details.

ChatGPT prompts for recruiters

Good prompts give ChatGPT the role facts, the audience, and the format, then ask for one output. Each prompt below covers a daily task in ChatGPT for recruiters, plus what to check before you use the output.

Job post draft

Prompt: "You're a recruiter writing a job post for a [title] role at a [company type] in [location]. Use only these intake notes: [notes]. Include a short summary, key duties, must-have skills, and the pay range [range]. Use plain, inclusive language and no buzzwords."

Check: Match every duty, skill, and pay detail against your notes. Delete anything the hiring manager or client never said.

Boolean string

Prompt: "Build a boolean search string for a [title] using these alternate titles: [titles]. Require these skills: [skills], and exclude [terms]. Group synonyms with OR inside parentheses and join the groups with AND."

Check: Run the string and read the first page of results. Remove synonyms that pull in the wrong roles, like "account manager" showing up in a project manager search.

X-ray query

Prompt: "Write an X-ray search query that finds public [title] profiles in [city] on [website]. Use the site: operator, put exact phrases in quotes, and add [skills] as required terms."

Check: Confirm the site path matches how that website structures its profile pages. Test the query yourself, since a wrong path returns nothing useful.

Outreach message

Prompt: "Write a short first message to a [title] about a [role] at a [company type]. Mention why their background fits: [your notes on fit]. Keep it brief and friendly, and end with one clear question."

Check: Make sure every fit detail is true and specific to this person. Cut flattery that would read like a template.

Follow-up message

Prompt: "Write a polite follow-up to a candidate who hasn't replied to my message about a [role]. Add one new detail they didn't get before: [detail]. Keep it shorter than the first message."

Check: Confirm the new detail is accurate and approved for sharing with candidates before you send it.

Interview question bank

Prompt: "Create behavioral and skills questions for a [title] interview. Base them only on these must-haves: [skills]. For each question, add what a strong answer would include."

Check: Drop any question that touches age, family plans, health, or other personal topics unrelated to the job. Make sure every question maps to a listed must-have.

Notes summary

Prompt: "Summarize these intake notes into a one-page brief for the hiring team. Use headings for must-haves, nice-to-haves, pay, timeline, and open questions. Notes: [notes with names and identifiers removed]."

Check: Compare the summary with your notes line by line, and look for anything it softened, dropped, or overstated.

Where to stop: decisions and data that stay out

Stop where the output would decide something about a person or needs their personal data to work. Better prompts can't fix this, because the risk sits in the task, not the wording.

• Hiring decisions: Don't ask it to pick, rank, or reject candidates. You own those calls, and you need reasons you can explain to the hiring manager, the client, and the candidate.
• Candidate personal data: Keep names, contact details, resumes, and notes about real people out of a consumer chatbot. Use placeholders like [Candidate A], and check your company's approved tools before you paste anything.
• Claims about the role or company: Check every fact it writes about pay, benefits, location, team size, or company history. It can fill gaps with details that sound right and aren't.
• Biased wording: Read every draft for gendered terms, age signals like "digital native," and requirements the job doesn't need. Ask it for neutral wording, then read the result yourself.

A quick test helps here: if you'd be uncomfortable showing the prompt and the output to the candidate, keep that task out of the chat.

Check the data gap: ChatGPT vs recruiting tools

Use ChatGPT for writing and a recruiting tool for finding people, since a chatbot has no candidate database behind it. That difference decides where ChatGPT for recruiting fits in each step of a search.

In a 2024 article, SHRM noted that many generative AI apps are built for HR or recruiting work. Examples include apps that summarize interview notes and detect cheating on candidate assessments. Our AI recruiting guide covers how to judge those tools before you buy.

Before you pick a search tool, ask where its data comes from, whether candidates opted in, and how it explains each match.

Start with one task before you scale

Start with the task that eats the most of your week, and run one prompt on it for a few open roles. The easiest place to begin with ChatGPT for recruiting is repeat writing, like job posts and follow-ups, where reviewing a draft is faster than writing from zero.

Save the prompts that work in a shared doc so your team edits the same versions. Put the check step next to each prompt so new recruiters review output the way you do. Add a second task once the first one saves time without extra rework.