Is It Cheating to Use AI in a Job Interview?

There’s no one-word answer here, and anyone who hands you one is selling something. It’s one of the questions job seekers keep googling in 2026, and the honest reply is a framework, not a verdict.

The question worth asking isn’t “AI or no AI.” It’s whether you’re representing yourself accurately or deceiving the person across the table. That one distinction sorts almost every case. Here’s how it works, with the data, the recruiter quotes, and a spectrum you can use before your next interview.

What does “cheating” mean in an interview?

An interview tests you. Your judgment, your real experience, how you think out loud under a little pressure. Cheating is anything that fakes a result the test is supposed to measure. So the question for any AI tool is simple: is it helping you communicate who you genuinely are, or inventing a version of you that doesn’t exist?

Built In put it well in their roundup of recruiter opinion. The answer is “it depends,” and what it depends on is deception (Built In, 2026). Prep tools that sharpen a real skill are like studying. Tools that supply expertise you lack, in the moment, while the interviewer thinks you’re unassisted, are like smuggling the answer key into a closed exam. Same technology, opposite ethics. What separates them is whether you can back up what you said when the follow-up question lands.

Where exactly is the line between fine and cheating?

Picture a spectrum. On one end, AI that builds your skill and helps you show your real self. On the other, AI that fabricates a candidate who doesn’t exist. Most real situations sit cleanly at one end or the other. The table below maps it.

The same ethical test shows up in nearly every source. Is the AI helping you communicate your real experience, or manufacturing credentials? Using AI to surface your own work is closer to an open-book exam, since the knowledge is yours. Using it to invent a track record is forgery.

How many people use AI in interviews?

A large minority, and the number is climbing fast. In a survey of 3,617 verified professionals, 1 in 5 (20%) admitted secretly using AI during interviews, and 55% agreed it has become “the new norm” (Blind, 2025). A 2026 Resume Genius survey of 1,000 US job seekers put real-time interview use at 22%, with 78% using AI somewhere in the search (Newsweek, 2026).

The vendor data points the same way, with one caveat. Fabric, a company that sells cheating-detection software, analyzed 19,368 AI-powered interviews and flagged 38.5% of candidates for cheating behavior, with rates climbing from 9% to 45% between July and September 2025 (Fabric, 2026). Read that one as directional, since the source profits from the alarm. But across independent surveys and vendor analytics alike, nobody disputes the trend. Live AI use in interviews went mainstream in roughly eighteen months.

What do employers and recruiters think?

Most of them treat hidden, real-time AI use as cheating, full stop, and they’re getting better at catching it. In a Checkr survey of 3,000 US hiring managers, 59% had personally suspected a candidate of using AI to misrepresent themselves, and 62% agreed job seekers are now better at faking with AI than hiring teams are at detecting it (Checkr, 2025). Greenhouse found 65% of hiring managers had caught applicants using AI deceptively, and 91% of recruiters had spotted some form of candidate deception (Greenhouse, 2025).

Companies are responding with policy. Amazon told recruiters that candidates may be disqualified for using AI tools in interviews, calling it an unfair advantage. As industry analyst Ian Silvera put it, “If you want to look like a flesh-bound chatbot, then by all means use an AI teleprompter” (IT Pro, 2025). Google, Cisco, and McKinsey all brought back in-person rounds specifically to counter AI cheating, and Gartner data shows 72.4% of recruiting leaders now run in-person interviews to fight fraud (Computerworld, 2025).

Not everyone is banning it. Meta is testing “AI-Enabled Interviews,” and Canva replaced its computer-science fundamentals round with an “AI-Assisted Coding” interview in June 2025 (LockedIn AI roundup, 2026). Policy varies wildly by company, which is exactly why you have to ask instead of assume.

Does disclosure change whether it’s cheating?

Yes, more than any other single factor. Most of what looks like cheating is really about hiding, and the same behavior that gets you disqualified when concealed can be fine when the employer knows and agrees. That gives you a clean self-test. Would you be comfortable saying it out loud? “I prepped with an AI coach and built a story bank” passes easily. “I’m reading answers off a hidden overlay right now” does not. There’s a fairness angle too. 57% of candidates think companies should be required to disclose when AI is judging them (Greenhouse, 2026), and the honest, symmetric standard is to be just as open about AI helping you. When and how to actually bring it up is its own decision, which we cover in our guide on whether to tell your interviewer you use AI.

Surfacing your experience vs. feeding you answers

This is the whole ballgame, and one real follow-up question settles it. When AI reminds you of a project you genuinely led, the interviewer can dig in (“walk me through the 2 a.m. production failure”) and you’ve got the war stories to answer. When AI hands you expertise you don’t have, the same probe exposes you, because there’s nothing underneath the script.

That failure mode shows up in the data. In the Resume Genius survey, 36% of job seekers admitted listing skills they don’t possess (Newsweek, 2026). Recruiters describe what it looks like in the room. Marshall Scabet of Precision Sales Recruiting recalled a candidate using AI to answer questions: “It was apparent she did not have the experience she said she did.” Cristina Flaschen of Pandium described “reading something scrolling on the screen that I can see in the reflection of their glasses” (Built In, 2026). Matt Erhard of Summit Search Group put the principle plainly: recruiters want “the candidate’s answers in their own words.”

So the distinction isn’t philosophical, it’s operational. AI that points you back to your own material survives scrutiny. AI that invents a fake you collapses on the first real question.

Where does a tool like Cluely fit?

Cluely is the example this whole debate keeps circling back to, and it sits at the far end of the spectrum. It grew out of Interview Coder, a tool whose founders fed AI answers through an “undetectable” overlay during live coding interviews and got suspended from Columbia in 2025 (CNBC, 2025), then relaunched with “Cheat on Everything” marketing (TechCrunch, 2025). The whole product is built on hiding from the interviewer rather than helping you show your real self. That’s the cheating end, plain and simple, and several rivals are now scrubbing the word “cheating” from their pages while keeping the stealth feature. For a full breakdown of these tools, including pricing and privacy track records, see our comparison of the best AI interview tools.

How can you use AI in a way you’d be comfortable disclosing?

Keep AI in prep, keep yourself in the answers, and only do live what you’d say out loud. The best model here is, fittingly, Anthropic’s own candidate guidance. After briefly banning applicant AI use in May 2025, the company reversed in July to a stage-by-stage policy (Fortune, 2025): use Claude to refine a resume you drafted yourself, skip it on take-home assessments unless told otherwise, and on live interviews, “this is all you, no AI assistance unless we indicate otherwise” (Anthropic candidate guidance).

Five rules that follow from all of this:

1. Prep with AI freely. Practice questions, company research, refining how you describe real wins. Universally accepted.
2. Default to no real-time AI in a live conversation unless the employer explicitly allows it. Check the policy, and ask if it’s unstated.
3. Use the disclosure test. Only do what you’d be comfortable saying to the interviewer’s face.
4. Never let AI claim experience you can’t defend under follow-up. Interviewers probe for war stories precisely to expose this.
5. Avoid hidden, “undetectable” overlay tools. They’re built to deceive, and getting caught is reputationally serious.

For a deeper prep playbook (prompts, story banks, company research), see how to prepare for a job interview with AI.

Where does Hedy stand on this?

We built Hedy on the ethical side of the line on purpose, and we’ve turned down features that would put it on the other. Hedy is a real-time conversation coach that points you back to your own resume and experience. It surfaces the achievement that fits the question, reminds you of the metric you forgot, and suggests sharp questions to ask the interviewer. What it deliberately won’t do is feed you scripted answers or pretend to be invisible.

People have asked us to build “make it undetectable” and “auto-answer the question” features. We refuse. That’s the exact opposite of a tool like Cluely, whose entire launch was built around cheating. Our position: an interview should still measure you, and a good coach makes you sharper at being yourself, not better at being someone else.

In practice, that means three legitimate uses across a job search. First, prep. Feed Hedy the job description and your resume and rehearse out loud while it coaches. Second, live coaching in virtual interviews, pointing you to your own material where policy allows it. In person, you won’t stare at your phone, so the value there is prep and review. Third, cross-round intelligence. Group every conversation with one company into a Topic, and Hedy connects what each round reveals. A job seeker might go through several rounds before she meets the two owners. Across those calls, Hedy can build up a read on what each owner cares about, then coach her on which parts of her background to emphasize with each one. That’s preparation, not deception, and it runs on-device, so your interview conversations stay private. There’s more on the AI interview tool for job seekers page.

The honest bottom line

Using AI in a job interview is cheating when it deceives, and it’s good practice when it doesn’t. Prep with it as much as you want. Let it help you remember and articulate your real experience. Disclose when the situation calls for it, and never lean on it to claim things you can’t defend. Do that, and you’re not gaming the interview. You’re showing up as the most prepared, most articulate version of yourself, which is the entire point.

Frequently asked questions

Is it cheating to use AI to prepare for an interview?

No. Using AI to generate practice questions, run through answers out loud, research the company, or get feedback on draft answers is widely accepted, equivalent to hiring a coach or using prep books. Even strict employers like Anthropic explicitly endorse using AI to refine application materials you drafted yourself.

Can employers tell if you’re using AI in an interview?

Increasingly, yes. Interviewers watch for tells: unnatural answer delays, eyes tracking a second screen, and explanations that don’t match the answer produced. In a Checkr survey, 59% of hiring managers had suspected a candidate of using AI to misrepresent themselves. Follow-up questions that probe for specifics are the most reliable way they catch scripted answers.

Is it illegal to use AI during a job interview?

As of mid-2026, no jurisdiction has made it illegal for a candidate to use an AI assistant in a standard interview. This is not legal advice and rules can change. Note that proctored third-party assessments (HireVue, HackerRank, LeetCode-style) almost always forbid outside aids, and violating those terms carries real consequences.

Do you have to tell an interviewer you used AI?

There’s no universal rule, but the practical standard is this: treat undisclosed help as cheating unless the employer explicitly allows it. Prep generally needs no disclosure. Live, in-the-moment AI usually does. The clean self-test is whether you’d be comfortable saying it out loud. We cover this fully in our disclosure guide.

Is using AI in a coding interview cheating?

It depends entirely on the employer’s policy. Anthropic, Google, and Amazon prohibit it. Meta and Canva now invite candidates to use AI in coding rounds because the job itself is AI-assisted. Using a hidden overlay to solve problems you couldn’t solve unaided, while the interviewer believes you’re working unassisted, is cheating in any context that hasn’t explicitly allowed it.

What is Cluely, and is it cheating to use it in interviews?

Cluely is a desktop tool that feeds live AI answers through an “undetectable” overlay during interviews and meetings. It grew out of Interview Coder, which got its founders suspended from Columbia in 2025. Using it to read AI answers while presenting them as your own thinking is deception, and it’s the clearest example of the cheating end of the spectrum.

Why did Anthropic ban then allow AI in its hiring process?

Anthropic banned applicant AI use in May 2025, then reversed in July to a stage-by-stage policy. Banning was hard to police, and AI prep “levels the playing field.” The current rule: use Claude to refine a resume you drafted, skip it on take-homes unless told, and keep live interviews AI-free unless indicated. It’s a good model of a thoughtful, disclosed policy.

How many candidates use AI during job interviews?

A 2025 Blind survey found 20% of professionals admitted secretly using AI during interviews, and a 2026 Resume Genius survey put real-time use at 22%, with 78% using AI somewhere in their search. A cheating-detection vendor, Fabric, flagged 38.5% of 19,368 interviews for cheating behavior, though that figure should be read as directional given the source.

What do recruiters think about candidates using AI in interviews?

Most treat hidden, real-time use as fraud and prep use as fine. Recruiters report catching candidates “reading something scrolling on the screen.” Several major companies (Google, Cisco, McKinsey) reinstated in-person rounds to counter AI cheating, while a minority (Meta, Canva) now invite AI in. They want, in one recruiter’s words, “the candidate’s answers in their own words.”

Is using AI to write your resume cheating?

Writing your resume entirely with AI, including fabricated achievements, crosses into misrepresentation. Using AI to polish wording after you draft it yourself, with your real experience, is something most employers treat as fair game, and strict ones even encourage it. The line is the same as everywhere else: refine the truth, don’t manufacture it.

Will I be disqualified for using AI in an interview?

You can be, if you use it deceptively. Amazon tells recruiters candidates may be disqualified for using AI tools in interviews. The risk concentrates on hidden, real-time use and fabricated credentials. Prep use carries no real disqualification risk, and AI use the employer has explicitly invited carries none either.

Does using AI to ask better questions count as cheating?

No. Asking sharp, well-researched questions of the interviewer is something candidates have always prepared for, with or without AI. Whether you brainstormed those questions with a coach, a friend, or an AI tool beforehand, they’re still your questions, asked in your own voice. This is one of the safest and most useful ways to bring AI into your interview prep.