51% of organizations now use AI to support recruiting, and AI use across HR climbed to 43% in 2025 from 26% a year earlier, per SHRM's 2025 Talent Trends research.Source ↗
Writing job descriptions is the top recruiting use of AI at 66%, ahead of screening resumes (44%), automating candidate searches (32%), and communicating with applicants (29%).Source ↗
37% of talent-acquisition professionals are experimenting with or integrating generative AI, and users report a 20% reduction in weekly workload — about a day a week, per LinkedIn's Future of Recruiting 2025.Source ↗
89% of HR professionals whose organization uses AI in recruiting say it saves time or boosts efficiency and 36% report lower hiring costs, yet 67% say their org has not proactively trained staff to work with AI.Source ↗
writingClaudeChatGPT

Writing an inclusive, skills-based job description

Job descriptions are the single most common AI use in recruiting (66% in SHRM's 2025 data), and the reason is simple: the boilerplate ones repel good people. Inflated "requirements," an endless wish list, and quietly gendered language shrink and skew the applicant pool before anyone applies. AI is good at a fast first draft you then tighten around what the role actually needs.

Prompt
You are helping an experienced recruiter draft an inclusive, skills-based job description. You write the first draft; I edit it to match the real role.

Role and level: {{role_title}}
What the person actually needs to do the job (skills and outcomes, not credentials): {{must_have_skills}}
Team, mission, and who they report to: {{team_context}}
Location, work model, and anything I can state about pay: {{comp_and_location}}

Write a job description with: a two-sentence hook, a "What you'll do" list of 5-6 outcomes, a "What you'll need" list split into must-haves and nice-to-haves, and a short "How we work" paragraph.

Rules:
- Base every requirement on the skills and outcomes I gave you. Do NOT invent responsibilities, technologies, years-of-experience bars, or benefits I did not state — insert [VERIFY] where a detail is missing.
- Keep must-haves short; move anything not truly required to nice-to-haves. Long requirement lists narrow and skew the pool.
- Flag any wording that could read as gendered, ageist, or exclusionary ("rockstar," "young and energetic," "native English speaker") and suggest a neutral alternative.
- Plain language, active voice, no superlatives. Aim for under 500 words.

Fill in your details and the prompt updates live — then copy.

communicationClaudeChatGPT

Personalized outreach that gets replies

Cold outreach is where most sourcing dies — generic InMails get ignored, and passive candidates can smell a mail-merge from the subject line. Communicating with candidates is one of the top AI uses in recruiting, and LinkedIn's 2025 data ties heavy use of AI-assisted messaging to a higher rate of quality hires. AI is good at turning a few real details into a short, specific note — as long as the details are real.

Prompt
You are helping a recruiter write a short, specific outreach message to a passive candidate. The goal is a reply, not a hard sell.

One real, public detail about the candidate I want to reference (I will type this myself): {{candidate_hook}}
The role and the honest one-line reason it is interesting: {{role_pitch}}
Why this person specifically — the skills match: {{why_them}}
Channel and length limit: {{channel_and_length}}

Write two versions of the message. Each must:
- Open with the specific detail I gave you, not a generic compliment.
- Say what the role is and why it might fit them, in one or two sentences.
- Make one clear, low-pressure ask (a 15-minute call, or "worth a conversation?").
- Sound like a person: contractions, short sentences, no corporate throat-clearing, no "I came across your profile."

Rules: use ONLY the details I provided. Do not invent facts about the candidate, their tenure, or their accomplishments, and do not claim the role offers pay, title, or perks I did not state. If something is missing, use a [FILL] placeholder. Keep each version inside the channel limit.

Fill in your details and the prompt updates live — then copy.

automationClaudeChatGPTCopilot

Boolean search strings for hard-to-find candidates

Finding the right people in LinkedIn Recruiter, an ATS, or a Google X-ray means writing Boolean strings full of title variants, skill synonyms, and exclusions — tedious to build and easy to get subtly wrong. Automating candidate search is a core AI use (about 32% of teams), and this is a task AI does genuinely well: it generates and explains the strings so you can tune them by hand.

Prompt
You are a sourcing expert who writes and explains Boolean search strings for a recruiter. I will describe the role; you give me strings I can tune.

Role and core skills: {{role_and_skills}}
Must-haves vs. nice-to-haves: {{must_have_vs_nice}}
Where I am searching (LinkedIn Recruiter, ATS, Google X-ray, etc.): {{platform}}
Titles, companies, or terms to exclude: {{exclusions}}

Return:
1. Three Boolean strings — one tight (high precision), one broad (high recall), and one in the middle — ready to paste, using AND / OR / NOT and quotation marks correctly for the platform I named.
2. A plain-English note on what each string will over- or under-match, so I know which to reach for.
3. A synonyms-and-variants list: alternate job titles, adjacent skills, and common misspellings I should consider adding or removing.

Rules: tailor the syntax to the platform I gave you and flag anything that platform does not support (LinkedIn ignores parentheses; a Google X-ray needs a site: operator). Do not claim a string will find a specific number of people or specific individuals — these are starting points I will test and refine. If the role is too vague to search well, tell me exactly what to specify.

Fill in your details and the prompt updates live — then copy.

planningClaudeChatGPT

A structured interview guide and scorecard

Unstructured interviews are where bias creeps in and where two interviewers rate the same candidate wildly differently. Structured, job-related questions with a shared scorecard are the best-evidenced way to make hiring both fairer and more predictive — and skills-based assessment is a stated 2025 priority for talent teams. AI is good at turning a role's real competencies into consistent questions and rating anchors.

Prompt
You are an interviewing expert helping a recruiter build a structured, job-related interview guide and scorecard. Consistency and fairness matter more than cleverness.

Role and the 3-5 competencies that actually predict success in it: {{role_and_competencies}}
Which stage this interview is (recruiter screen, hiring manager, panel, etc.): {{interview_stage}}
Time limit and format: {{format_constraints}}

Produce:
1. For each competency, one open behavioral question ("Tell me about a time...") plus two follow-up probes that get past a rehearsed answer.
2. A scorecard: each competency scored 1-4, with a one-line behavioral anchor describing what a 1, 2, 3, and 4 look like, so different interviewers rate consistently.
3. A short note on what to genuinely listen for versus what is just a good talker.

Rules: every question must be job-related and tied to a competency I gave you — do NOT invent competencies or add questions unrelated to the role. Do NOT generate any question that asks about or reveals age, race, religion, national origin, disability, pregnancy, marital or family status, or other protected characteristics; if a competency I listed edges into that territory, flag it instead of writing the question. Keep the whole guide inside the time limit I set.

Fill in your details and the prompt updates live — then copy.

analysisClaudeChatGPT

A job-related screening rubric you apply, not the AI

Resume screening is the second-most-common AI use in recruiting (44%) and also the most legally loaded. Letting a chatbot rank or reject candidates from their resumes is exactly what triggers bias-audit laws and disparate-impact liability. The safe, useful version flips it: have AI build a job-related scoring rubric from the role and give you a consistent way to record your own assessments — you do the judging.

Prompt
You are helping a recruiter build a fair, job-related screening rubric for a role. Important boundary: you design the rubric; you do NOT evaluate, rank, score, or reject any actual candidate, and I will not paste resumes or candidate data into this chat.

The role's real requirements and must-have skills: {{job_requirements}}
How I currently decide who moves forward — my rough criteria: {{screening_criteria}}

Produce:
1. A screening rubric that turns each must-have into an observable, job-related signal ("has shipped X" rather than "good communicator"), with a simple pass / maybe / no-signal scale and a weight showing which requirements matter most.
2. A short, consistent template I can use to write down why each candidate did or did not meet each criterion, so my own notes stay structured and comparable.
3. A "watch out" list: criteria in my rough notes that are risky proxies for protected characteristics (graduation year, employment gaps, "culture fit," school prestige) and a job-related signal to use instead.

Rules: build the rubric only from the requirements I gave you — do not invent must-haves. Do not offer to screen resumes or rate candidates for me; the rubric is a tool I apply by hand. Keep it plain and auditable.

Fill in your details and the prompt updates live — then copy.

creativeClaudeChatGPTGemini

Recruitment-marketing content for a hard-to-fill role

A job post alone rarely fills a competitive role, so recruiters increasingly write social posts, "why join us" blurbs, and campaign hooks to build a pipeline — and employer branding is a headline theme in LinkedIn's 2025 report. AI is good at spinning one set of real selling points into several formats, so you are editing rather than starting from a blank page every time.

Prompt
You are a recruitment-marketing writer helping a recruiter promote a specific open role. Honesty and specificity beat hype.

The role and its genuine selling points — real projects, growth, team, mission, things I can stand behind: {{role_and_selling_points}}
Who I am trying to reach and what they care about: {{audience}}
Where this will run and the tone I want: {{channels_and_tone}}

Produce:
1. A LinkedIn post under 150 words that leads with the most compelling real selling point, not the job title.
2. A short "Why join this team" paragraph I can reuse in outreach and on the careers page.
3. Three subject lines or hooks that take different angles on the same role.

Rules: use ONLY the selling points I gave you. Do NOT invent perks, salary, culture claims, awards, or benefits — if a post feels thin, tell me what real detail would strengthen it rather than making one up. Avoid clichés ("rockstar," "work hard play hard," "we're a family") and anything that reads as age- or gender-coded ("digital native," "young team"). No emojis unless I ask. Keep every claim something a candidate could hold us to on day one.

Fill in your details and the prompt updates live — then copy.

Common questions from recruiters

Is it legal to use AI in hiring?

Yes, but hiring is heavily regulated. Title VII disparate-impact liability applies to AI-driven selection, and you are responsible even when a vendor's tool caused the bias. New York City's Local Law 144 requires an independent bias audit and candidate notice for any tool that scores or ranks candidates, and Illinois and Colorado have AI-hiring laws taking effect in 2026. Using AI to draft is low-risk; using it to decide is where the law bites.

Didn't the EEOC drop its AI rules in 2025?

The EEOC removed its non-binding AI technical-assistance documents in January 2025, and a federal executive order told agencies to deprioritize disparate-impact enforcement. But those documents only explained existing law — they did not create it. Title VII, the ADA, and state and local laws like NYC Local Law 144 are all still in force and, at the state level, actively expanding.

Can I paste a candidate's resume into ChatGPT to screen it?

Not into a consumer tool. A resume is candidate PII covered by GDPR, CCPA, and your firm's data obligations, and consumer chatbots may retain or train on what you paste. A tool that scores or ranks candidates is also likely an AEDT that requires a bias audit and advance notice. Use AI to build a job-related rubric, then apply it to real candidates yourself.

Where should a recruiter start with AI?

Start with the writing that carries no candidate data: job descriptions, outreach templates, Boolean strings, and interview guides. Job descriptions are already the top AI use at 66%, and these tasks pay off in the first week without touching a candidate's personal information or any hiring decision.

Related professions