The job description is the first interaction a candidate has with your organisation. It is also, in most enterprises, an afterthought: a copy-paste from the last time the role was hired, updated with the new manager's wish list, reviewed by nobody for language quality, and posted. The result is a document that actively discourages the candidates you most want to attract. AI can fix this systematically.
The Problem With Job Descriptions
I have reviewed hundreds of job descriptions across my career. The patterns of failure are consistent.
Gendered language. Research consistently shows that words like "dominant," "competitive," and "aggressive" discourage women from applying, while words like "collaborative," "supportive," and "nurturing" discourage men. Most job descriptions use gendered language unconsciously. The hiring manager did not intend to exclude. But the effect is the same regardless of intent.
Credential inflation. "Must have 10 years' experience" for a role that a capable person could perform after 3. "MBA required" for a role where an MBA adds no value. Credential inflation narrows the applicant pool without improving candidate quality. It disproportionately excludes career changers, people from non-traditional backgrounds, and younger candidates who are fully capable but credential-poor.
Vague expectations. "Must be a team player." "Strong communication skills required." "Results-oriented." These phrases mean nothing specific. They fill space without communicating what the role actually involves or how success will be measured.
Kitchen-sink requirements. A 15-item list of requirements that no single human being possesses. The best candidates read the full list, assess honestly that they meet 70% of it, and decide not to apply. The worst candidates read it, assess that they meet 40%, and apply anyway. The requirements list inverts the selection effect you want.
44%
of women will not apply to a role unless they meet 100% of listed requirements, compared to 16% of men
Source: LinkedIn, Gender Insights Report, 2024
How AI Fixes This
Language Analysis
AI scans the job description for gendered language, exclusionary phrasing, and tone problems. It identifies specific words and phrases that research shows discourage certain demographics from applying, and suggests neutral alternatives.
This is not political correctness. It is recruitment effectiveness. A job description that inadvertently discourages 44% of female candidates from applying has a 44% smaller talent pool. In NZ's tight labour market, that is a competitive disadvantage.
Requirements Calibration
The AI analyses the requirements list against role benchmarks: what do people actually performing this role typically have? This surfaces credential inflation. "10 years' experience required" when the role benchmark is 3-5 years. "Degree required" when the role benchmark shows successful performers with diplomas and industry certifications.
The AI does not remove requirements. It flags where your requirements exceed the benchmark and asks whether the inflation is intentional or inherited from the last posting.
Clarity Enhancement
Vague phrases get replaced with specific descriptions. "Strong communication skills" becomes "writes clear internal reports and presents findings to senior stakeholders monthly." "Team player" becomes "collaborates with a cross-functional team of 6 to deliver quarterly projects."
Specific descriptions help candidates self-select accurately. They also help hiring managers align on what they are actually looking for, which improves interview quality and reduces mis-hires.
Structure Optimisation
AI restructures the job description for maximum applicant engagement. Research shows that the order of information matters: role purpose and impact first, key responsibilities second, requirements third, benefits fourth. Many job descriptions put requirements first and purpose last, which is exactly backwards from what motivates the best candidates.
Loading demo...
What I Have Learned About Adoption
The adoption pattern for AI job descriptions is interesting. HR teams are enthusiastic. Hiring managers are resistant. The resistance comes from a reasonable place: hiring managers feel they know their team and their requirements better than an AI.
The conversation that works is not "the AI will write better job descriptions than you." It is "the AI will ensure your job description reaches the widest possible pool of qualified candidates." The former threatens competence. The latter offers leverage.
The practical approach:
- Hiring manager writes the initial brief (requirements, responsibilities, context)
- AI generates the job description with inclusive language, calibrated requirements, and clear structure
- Hiring manager reviews and adjusts
- AI flags any changes that reintroduce gendered language or credential inflation
- Final review and posting
This keeps the hiring manager in control while ensuring the output meets quality standards.
The Diversity Connection
Inclusive job descriptions are necessary but not sufficient for diverse hiring. A perfectly written job description posted only on the channels where your existing team found their roles will reach the same demographic.
But the job description is the first filter. If it excludes before the process even begins, nothing downstream matters. Getting the job description right is the highest-leverage single intervention in the recruitment pipeline.
Implementation
For HR teams ready to improve their job descriptions:
- Baseline audit (1 week). Analyse your current job descriptions for gendered language, credential inflation, and clarity issues. This establishes the improvement opportunity.
- AI configuration (1-2 weeks). Configure the generation model with your organisation's tone, values, and role frameworks.
- Pilot (2-3 weeks). Use AI-assisted generation for 5-10 open roles. Compare applicant pool diversity and quality against historical baselines.
- Rollout (ongoing). Integrate into the standard recruitment workflow.
Total: 4-6 weeks to establish the pattern. The ongoing effort is minimal because the AI handles the heavy lifting for each new role.
Better job descriptions attract better candidates. Better candidates make better teams. Better teams deliver better outcomes. It starts with the words.