An AI Interviewer Hung Up on a Candidate. Are AI Job Interviews Fair?

This week the internet found a new villain: an AI interviewer called "Sarah".

In a TikTok from creator The Dreadful Studio (@thedreadfulstudio), a candidate sits through an AI video screening for a management role and answers every question in made-up words like "zibo" and "shima". The avatar smiles while he talks, then scowls when he stops. Finally it snaps. "This is an absolute waste of my time," it says, before ending the screening with "You are completely unqualified. Goodbye" (Futurism, IBTimes UK).

By October 8 it had been picked up by Inc., Entrepreneur and Tech Times UK, and the comment sections had turned into a referendum on whether a bot should be allowed to interview you at all.

#Is the viral AI interview video real?

Nobody has verified it. No employer or AI vendor is named in any coverage. Inc. described the clip as "staged for laughs", Entrepreneur said it "appears" authentic, and Tech Times UK flagged it as unverified. The same creator has posted other AI interview videos, including one answering in Eminem lyrics (Inc.).

It is also not the first clip like this. In September, satirist Graham Zip fed an AI recruiter nonsense answers and the bot cheerfully accepted them (Futurism). One bot was too harsh, the other too easy. That contrast is the whole debate in miniature.

But whether or not this particular video is real almost doesn't matter. It went viral because it landed on something millions of job seekers have already felt.

#Why the clip hit a nerve

AI interviews are no longer a niche experiment. Greenhouse surveyed 2,950 job seekers across the US, UK, Germany, Australia and Ireland for its 2026 Candidate AI Interview Report (Greenhouse). The results:

• 63% had been interviewed by AI, up 13 points in six months.
• 70% weren't clearly told upfront that AI would evaluate them, and 21% only found out once the interview had started.
• After an AI interview, 51% never heard back. Only 28% advanced.
• 38% had abandoned a hiring process because of an AI interview, and another 12% said they would.

Greenhouse sells hiring software, so treat it as a vendor survey. Still, the pattern matches what people say online: you talk to a screen, nobody tells you how you are being judged, and then nothing happens.

#The case against AI job interviews

They feel dehumanizing. The most repeated reaction to the clip was a Reddit comment calling AI interviews "dehumanizing and dystopian" (quoted by Futurism and Entrepreneur). If a company won't put a person in front of you, many candidates read that as a signal about how they will be treated once hired.

They can shut people out. Commenters asked what happens to a candidate with a stutter, a strong accent or a bad connection (Tech Times UK, IBTimes UK). A human can adjust. A rigid scoring model may not.

They can scale old problems. Johnny Campbell, CEO of SocialTalent, put it bluntly to SHRM (SHRM):

AI doesn't fix broken interviewing. It scales it.

Greenhouse's own chief people officer, Sharawn Tipton, said something similar about bias: "AI isn't fixing bias, it's scaling it." Interestingly, candidates in the Greenhouse survey reported almost identical levels of age bias (36%) and race or ethnicity bias (27%) from AI and human interviewers. Neither side comes out looking clean.

Some may not even be real jobs. In New Zealand, job seeker Charlie Maclean started a petition to Parliament after a timed AI interview cut her off mid-answer and the company went silent. She suspects the role was a ghost job used to collect candidate data. Job boards SEEK and Trade Me say ghost jobs are not a material problem on their sites, and there is no hard evidence yet (1News).

#The case for AI job interviews

Volume. US candidates per open role on LinkedIn have doubled over four years (LinkedIn research, via Entrepreneur). Recruiters can't phone-screen hundreds of people for every role. Mason Swofford of Tenzo AI told SHRM, "Human time can't be scaled up, but AI time can be."

Consistency. Supporters argue every candidate gets the same questions and the same rubric, instead of depending on a tired recruiter's mood at 4pm. HackerRank CEO Vivek Ravisankar went further this week: "AI is way less biased than humans, if you tune it properly" (TechCrunch). Note the condition at the end of that sentence.

More people get a shot. When a recruiter can only screen 20 people, 20 people get screened. An AI first round can, in theory, give everyone who applies a chance to speak instead of being filtered out on a resume keyword.

Fraud. Employers are also fighting fake applicants. Gartner has predicted that by 2028, one in four job applicants could be fake (AARP). Structured AI screens are one of the tools companies are using to catch them.

And in fairness to "Sarah": a candidate who answers every question in gibberish probably should not advance. The tone was the problem, not the outcome.

#Both sides at a glance

#This week, AI moved deeper into hiring

The viral clip landed in the same week HackerRank showed off an AI interviewer that rolls the recruiter screen, the take-home test and the engineer follow-up into one AI-run interview (TechCrunch). Whatever the internet thinks of "Sarah", the direction of travel is clear: more of the first round is going to be run by software.

#What the law says (so far)

Rules are patchy and depend on where you live, and this is not legal advice, but a few exist already (Gibson Dunn):

• New York City (Local Law 144) requires a bias audit and notice to candidates 10 business days before an automated hiring tool is used.
• Illinois requires employers to give notice when AI is used in hiring decisions, from January 1, 2026.
• Colorado has a law promising an explanation after a rejection and human review on request, formally effective January 1, 2027, though enforcement is currently on hold in court.
• In the UK, Tech Times UK notes the Equality Act 2010 may require reasonable adjustments, such as an alternative format, for disabled candidates.

Candidates clearly want more. In the Greenhouse survey, 57% said disclosure should be legally required, and the safeguard people wanted most (46%) was the option to request a human interviewer.

#Our take

The clip is entertaining, but it is not evidence that AI interviews are broken. Gibberish in, rejection out is arguably the system working. The real problems are quieter and more common: candidates not being told AI is judging them, no human fallback for people who need one, and silence after the interview.

At the same time, pretending AI interviews will go away is not a strategy. If 63% of job seekers have already faced one, the practical question for you is not whether they are fair. It is how to do well in one when it is the only door open.

#How to pass an AI job interview

Career experts interviewed by the AP gave advice that applies to AI and human interviewers alike (AP via The Inquirer):

1. Research the role and company. "The more prepared you are, the easier it will be to tailor your responses," said Amanda Augustine of Careerminds. Most AI interviews score you against criteria taken from the job posting, so the posting is your study guide.
2. Practise out loud. "You need to practice out loud," said Priya Rathod of Indeed. Reading answers in your head is not the same as saying them to a camera with a timer running.
3. Be specific. Use the STAR method (situation, task, action, result) and real numbers. Rathod warned against "giving really vague answers", which is exactly what scoring rubrics penalize.
4. Check your setup. Test your microphone, camera, lighting and connection before you start. A frozen screen is not a reason the AI will forgive.
5. Don't let another AI answer for you. Using AI tools during the interview "can sometimes immediately disqualify you," Rathod said.
6. Ask for clarity. Many AI interviewers will repeat or rephrase a question if you ask.
7. Ask about the process. It is fair to ask whether AI will evaluate you and whether a human option exists. If you need an accommodation, ask for it in writing.
8. Watch for red flags. An intense recruitment drive from a company nobody has heard of, with no human contact at any stage, deserves a second look (Victoria University's Ben Walker, via 1News).

The candidates who struggle most in AI interviews are usually the ones who meet the format for the first time on the real call. A few practice rounds with a realistic interviewer, out loud and on a timer, take most of the shock out of it. And because these tools match you against the job description, a resume tailored to the posting gets you to that interview in the first place. If you are not getting that far yet, start with why you might not be getting interviews.