This guide provides general technical information. It is not legal, medical, tax or financial advice. Confirm requirements with a qualified adviser for your jurisdiction and implementation.
Why employers want AI screening
A single open role can attract hundreds of applications. Recruiters need to read them, compare them with the job, and respond to candidates. AI promises speed: parse resumes into structured data, rank by fit and flag missing information. The time savings are real, especially for high-volume hiring.
The risk is equally real. Hiring decisions affect people’s livelihoods and are regulated.
Where the risks come from
- Biased training data. A model trained on past hiring decisions can learn the preferences, and prejudices, of past hiring managers.
- Proxy variables. Postal codes, school names, employment gaps or even hobbies can correlate with protected characteristics.
- Opacity. If no one can explain why a candidate was ranked low, you cannot defend the process or fix it.
- Over-reliance. Reviewers tend to follow a ranked list, so the AI’s mistakes become decisions.
- Candidate data privacy. Resumes contain personal information that needs protection.
Rules and guidance to be aware of
Regulation in this area is changing quickly, and it varies by place. Examples to check:
- In the US, federal anti-discrimination law applies to hiring tools used by employers, and the Equal Employment Opportunity Commission has published guidance on automated tools. The long-standing “four-fifths” rule of thumb is often used to screen for adverse impact in selection rates.
- New York City requires bias audits and candidate notices for certain automated employment decision tools. Other states and cities have introduced or passed their own rules, such as Illinois on AI in video interviews.
- In Canada, human rights legislation applies to hiring, and privacy laws govern candidate data.
Have employment counsel confirm which rules apply to you. This is general information, not legal advice.
A safer way to use AI in hiring
| Practice | What it looks like |
|---|---|
| Write job-related criteria first | Skills, experience and certifications that the role truly requires |
| Use AI to organize, not decide | Extract and summarize evidence against the criteria, then let a person decide |
| Remove unnecessary signals | Avoid using names, photos, addresses or other irrelevant details in scoring |
| Keep a human decision | A trained reviewer makes every advance or reject decision |
| Test for adverse impact | Compare pass rates across groups regularly, and investigate gaps |
| Give notice and an alternative | Tell candidates about automated tools and offer a way to request review where required |
| Log decisions | Record criteria, outputs and who decided |
Questions to ask a screening vendor
- What data was the model trained on, and how was bias tested?
- Can you provide independent bias audit results where required?
- What inputs does the tool use to score candidates?
- Can reviewers see why a candidate was ranked as they were?
- How is candidate data stored, shared and deleted?
- Does the tool train on our candidates’ data?
Lower-risk uses of AI in recruiting
Many teams get value without automated rejection: scheduling interviews, answering candidate questions, summarizing interview notes, formatting resumes consistently and keeping candidates informed. These save time without handing the decision to a model.
FAQ
Is AI resume screening legal?
It can be, but it is regulated. Anti-discrimination laws apply to automated tools, and some jurisdictions require bias audits and notices. Check the rules where you hire, and get employment counsel to review your process.
Does AI resume screening discriminate?
It can, if it is trained on biased data or uses proxies for protected traits. Reduce the risk with job-related criteria, human decisions, regular adverse impact testing and careful vendor selection.
Should AI reject candidates automatically?
Most experts advise against fully automatic rejection. Use AI to summarize and organize, and have a trained person make the decision, especially for borderline or unusual candidates.
What can AI safely do in recruiting?
Scheduling, candidate communication, resume formatting, interview note summaries and answering applicant questions are lower-risk uses. They save time without making employment decisions.
Sources and further guidance
Platform features and requirements change. Check the linked primary documentation for your implementation.
TURN INSIGHT INTO ACTION
Let’s talk about your next step.
Share your workflow, your current systems and what you want to improve. We’ll help shape a practical way forward.
Discuss your project ↗