How to tell candidates you use AI in hiring

Tell candidates before the interview starts, in plain words: that AI is involved, what it does, what it doesn't do, who makes the decision, and how to ask for another option.

Why it's worth doing well

Candidates are wary, and you should assume so. In a Pew Research Center survey of 11,004 US adults run in December 2022, 66% said they would not want to apply to an employer that used AI to help make hiring decisions. Americans opposed AI making the final hiring decision by 71% to 7%.

The same survey has a useful twist. 47% said AI would be better than humans at treating all applicants similarly, and 15% said worse. People are open to consistency. What they don't want is a machine deciding their future without anyone looking. A good notice speaks to both.

What to say

A clear notice answers six questions a candidate is already asking:

Is this a real person or AI?
What exactly does the AI do?
What is kept, and who sees it?
Who makes the decision?
What will I be assessed on?
What if I can't do it this way?

When to say it

Say it early, and say it again at the moment it matters. Mention AI in the job posting or the invitation, so nobody is surprised. Then repeat it on the screen before the interview starts, and ask for consent there.

Don't bury it in a privacy policy. A link to a long document doesn't tell anyone what's about to happen. Some laws also set their own timing, as you'll see below.

Sample wording

Adapt this to what your tool actually does. Don't copy a promise you can't keep.

"The first round of this process is an interview with an AI interviewer. It will ask you the same questions we ask every candidate. Afterwards, AI will score a transcript of your answers against the criteria for this role, which are listed below. A person on our team will read your transcript and make the decision. The AI does not decide. If you need an adjustment or can't take part this way, contact us at the address below and we'll work out another option."

What the law asks for

Rules differ by place and change quickly. Here are four to know about. This isn't legal advice and it isn't a full list, so get your own advice for where you hire.

New York City: Local Law 144 covers tools that substantially assist or replace discretionary decisions, including at the screening stage. Where it applies, candidates who live in the city must get notice 10 business days before use, including the qualifications or characteristics assessed and how to ask for a reasonable accommodation. The tool also needs a bias audit within the year before use, with a summary made public.
Illinois: a law in effect since January 1, 2026 requires employers to give notice when they use AI in covered employment decisions, including recruitment and hiring. It also bars using zip codes as a proxy for protected classes. Check the state's latest rules on how notice must be given.
EU AI Act, transparency: Article 50, in application since 2 August 2026, requires providers to design AI systems that interact directly with people so those people are told they are dealing with AI, unless that's obvious.
EU AI Act, high-risk: Annex III lists AI used to evaluate candidates as high-risk. Under Article 26, deployers must tell people that such a system is being used on them, and must assign human oversight. Those duties apply to Annex III systems from 2 December 2027, after Regulation (EU) 2026/1744 moved the date.

Offer another way in

A candidate may have a hearing, speech or other condition that makes a voice interview hard. In the US, the Americans with Disabilities Act requires employers to provide reasonable accommodations in the hiring process unless that would cause undue hardship, and an employer can't refuse to consider someone because they need one. The EEOC says employers may ask all applicants whether they need an accommodation for the hiring process itself.

So name a real contact in your notice, and answer quickly. Decide in advance what your alternative is, such as typed answers or a call with a person.

Questions candidates will ask

Have an answer ready for each of these, and make sure your team gives the same one.

Will my accent count against me? Your answer should say what is scored and what isn't.
Is this being recorded? Say exactly what is kept.
Can a person review my result? Say who to ask.
Can I see or delete my data? Say how, and who handles it.

What Vettika does, and what it doesn't

Vettika tells candidates it's AI before they start and asks for their consent. They can switch to typing at any point. We keep the transcript, not a recording, and there's no video. People on your team make every decision. Unfinished interviews can be resumed for 48 hours through an emailed link.

That screen doesn't replace your own notice. Vettika has not had a bias audit, so you can't rely on us for one. If you hire in New York City, or anywhere else with rules like these, get your own advice first.

Sources

Common questions

Do I have to tell candidates I use AI?

In some places, yes, and the rules depend on where you hire and what the tool does. Telling them everywhere is the simplest policy.

Where should the notice go?

In the job posting or invitation, and again on the screen before the interview starts. A privacy policy link alone isn't enough to be clear.

What if a candidate refuses the AI interview?

Have an alternative ready, such as a call with a person. If the refusal relates to a disability, treat it as an accommodation request.

See question templates