Candidate AI Use in Interviews

Candidate AI use in interviews covers any way applicants rely on generative AI before or during the hiring process, from practicing answers to reading AI-written responses live on screen.

Whether a given use is acceptable depends on the employer's policy, which is why many HR teams now state their position in the interview invitation.

What Counts as Candidate AI Use?

SHRM has covered the debate over whether using ChatGPT on application documents counts as cheating. The answer varies by employer.

Whatever your stance, apply it evenly. A rule enforced against some candidates but not others creates fairness risk, and it weakens your candidate screening process.

How Common Is It?

A Gartner survey of 3,290 candidates found that 39% used AI during the application process, and Gartner's later research found 6% admitted to interview fraud.

A Fabric analysis of nearly 20,000 interviews, reported by SHRM Labs, found 38.5% of candidates showed signs of cheating, rising to 48% in technical roles.

Signs of cheating are not proof, and the share depends on how each vendor defines it. Treat the numbers as direction, not a benchmark.

Candidates themselves are divided. Some see AI help as a normal productivity tool, while others see it as a way to hide what they cannot do. Employers have to decide where their line sits for each role.

How Can Interviewers Spot Real-Time AI Help?

• Strong initial answers that collapse when asked to explain their reasoning
• Exercises finished far faster than expected
• Scenario-based answers scoring lower than multiple-choice ones

Ask follow-up questions tied to the candidate's own answers, and record ratings in an interview scorecard so judgments stay consistent.

Interviewers can also change the format. Asking a candidate to talk through a past decision, then probing one detail, is hard to fake and easy to compare across applicants.

Calibrate across interviewers. If one person flags AI use at every turn and another never does, results become inconsistent and hard to defend.

How Should HR Write an AI Use Policy?

State what is allowed, such as preparation, and what is not, such as live answer tools. Gartner recommends explaining acceptable use up front.

Redesign assessments so knowing the answer is easier than gaming the test. SHRM Labs points to scenario-based tasks linked to real job work. Protect candidate experience by keeping rules short and specific.

Review the policy yearly. Tools and norms are moving fast, and a rule written for text chatbots may not cover live voice assistants or screen overlays.

See how HR Cloud's AI hiring agent supports candidates through applications, and review responsible AI use in HR for policy ideas.

Discover how our HR solutions streamline onboarding, boost employee engagement, and simplify HR management

Request a Demo

Frequently Asked Questions

Q: Is it cheating if a candidate uses AI to prepare?

A: Usually not. Practicing answers and researching the company are widely accepted. Live answer tools are the common line employers draw.

Q: Can employers ban AI use in interviews?

A: Yes, if the rule is stated clearly beforehand. Put it in the interview invitation and the assessment instructions.

Q: How do you detect AI-generated answers?

A: Look for gaps between written or spoken answers and the candidate's ability to explain them, unusual speed, and weak performance on open scenarios.

Q: Should AI detection software be used?

A: Treat detection as a prompt for follow-up questions, not a verdict. False positives can harm honest candidates.

Q: Does this affect remote interviews more?

A: Yes. Screen-sharing and overlay tools are harder to spot on video, so structured follow-up matters more.

Q: How should recruiters handle a suspected case?

A: Ask a follow-up in the same conversation, document what you saw, and apply the same process to every candidate.