100×worker · task analysis

Can AI write job ads?

By Hendrik De Winne Last updated: August 10, 2026 Lees dit in het Nederlands

Partly Yes, partly. AI drafts a solid first version of a job ad, covering structure, responsibilities, requirements, and benefits. You still need to check the text for neutral and non-discriminatory language, match the job title to your internal classification, confirm salary and benefits data, and verify that the description actually reflects the role.

How to do it

Start by gathering the raw material yourself: job title, task list, requirements, benefits, and team context. Give the AI an existing job ad from your company as an example of tone and style, then ask it to write a first draft that includes a pitch, a task list, a candidate profile, an offer, and an application procedure. This gives you a complete structure in minutes instead of a blank page.

This works well because writing a job ad is a structured, pattern-based text task, the kind of work where AI tends to perform strongly. Microsoft's Working with AI study, based on 200,000 real Copilot conversations, found that writing-heavy occupations score highest on AI applicability (translators topped the list at 49%). Anthropic's Economic Index adds an important nuance: most real-world AI writing use is augmentation, not full automation. The model produces a draft; a person still edits, checks, and approves it before it goes live.

Step by step

Collect the building blocks: job title, tasks, requirements, benefits, and team context.
Give the AI an existing job ad from your company as a style and tone reference.
Ask the AI to draft a full version: pitch, task list, profile, offer, and application process.
Check the draft for gender-neutral and non-discriminatory language.
Compare the job title and requirements against your official job classification (for example ESCO, or your internal grading system).
Create channel-specific versions for LinkedIn, Indeed, a national job board, or your internal careers page.

Where it breaks down

AI does not know your internal job grading system or your company's official title for a role, and it may suggest a job title that does not match your classification. It also does not know what salary or benefits are realistic for your sector, region, or company size: you have to supply and verify those numbers yourself.

The biggest risk is discriminatory or non-neutral language. Many jurisdictions require job ads to avoid gendered wording and to not imply age, background, or other protected characteristics (phrases like "young and dynamic team" are a common example). AI models sometimes produce this kind of language without flagging it. Read every draft before publishing it, and if you are unsure whether a phrase is compliant in your country, have someone from HR or legal review it. Legal requirements for job ad language differ by country and are not something a general-purpose AI model tracks reliably.

A prompt to start with

Draft a job ad

Write a job ad for the role of [job title] at [company name], a [industry/sector] company with [team size/context]. Use this tone and style as a reference: [paste an example of an existing job ad or describe your tone]. Include: a short company/team pitch, a task list (5-7 bullet points), a candidate profile (must-have and nice-to-have requirements), the offer (salary range, benefits, working arrangements: [insert details]), and a short application procedure. Use gender-neutral, non-discriminatory language throughout. Keep it under [word count] words.

Jobs where this task is shifting

Which of your tasks can AI take over?

The free scan builds your personal task map in 30 seconds.

Run the free task scan

Frequently asked questions

Can AI guarantee that my job ad is legally compliant?

No. AI can avoid obviously biased phrasing if you ask it to, but it does not know the specific anti-discrimination or labor law requirements of your country or region, and those requirements vary widely. It also cannot confirm that your salary range, benefits wording, or contract terms meet local legal standards. Treat any AI draft as a starting point, and have someone with legal or HR knowledge of your jurisdiction review the final text before publication.

Will AI automatically use gender-neutral language?

Not automatically, and not reliably. AI models are trained on large amounts of text that often contains gendered or biased phrasing, and they can reproduce it (for example, describing a team as "young and dynamic") without recognizing the problem. If you explicitly instruct the model to write in gender-neutral, non-discriminatory language, it usually improves the draft, but you still need to read the final text yourself before it goes live.

Can AI match my job title to an official occupational classification?

Partly. Standardized classifications like ESCO, which describes over 3,000 occupations across many languages, exist precisely because job titles vary a lot between companies and countries. AI can suggest a plausible standard title, but it does not automatically know which official classification your company or country uses, or which internal grading level a role belongs to. Verify the suggested title against your own system before publishing.

Sources

This article was drafted with AI assistance and editorially reviewed.