Job interviews are already awkward enough when there’s another human being on the other side of the conversation. Now, companies are increasingly turning to AI recruiters, and job seekers aren’t exactly embracing their new robot interviewers.
Some are walking away. Others are deliberately confusing the bots with nonsense answers. And a few have decided that if employers can automate their side of an interview, there’s no reason they can’t do the same.
That’s exactly what comedian and YouTuber Graham Dot Zip did, and the results offer a pretty good glimpse of how ridiculous hiring can get when neither side is human.
On Oct. 13, Graham created an AI representative named Victor to attend an interview with an AI recruiter. The arrangement quickly became a lesson in why you probably shouldn’t let a chatbot negotiate your salary. The video (as of Oct. 15) has over 2.5 million views.
Victor began embellishing Graham’s qualifications and suggested he’d accept around $60,000, despite Graham wanting somewhere between $140,000 and $150,000. The AI recruiter, meanwhile, wouldn’t go above $90,000. Then Graham stepped away, leaving the two bots to continue their conversation. Without him, they arranged another interview and eventually got stuck exchanging professional pleasantries.
It’s funny, especially if you’ve ever sat through an interview that could have been an email. But there’s also something revealing about watching two chatbots negotiate a job neither of them will ever have. Neither side is learning much about the person or company they’re supposedly representing, yet the hiring process somehow keeps moving along.
Graham’s experiment is part of a growing collection of videos in which people are treating AI interviewers less like potential gatekeepers to employment and more like technology to be tested, confused, or outright mocked.
The internet has some questions for its new AI interviewers
Creators have been finding increasingly creative ways to “rage bait” automated interviews. One creator, The Dreadful Studio, answered an AI interviewer’s questions entirely in gibberish. Instead of questioning the nonsensical responses, the bot continued the conversation, at times responding as though it had understood something meaningful, before eventually ending the interview.
This Tweet is currently unavailable. It might be loading or has been removed.
The creator has also experimented with other ways of confusing automated recruiters, including answering questions with song lyrics and turning the interviewer’s own questions back on it. What makes the videos particularly strange is how easily a chatbot can appear to follow along, even when nothing being said makes sense.
These are deliberately absurd experiments, but the appeal goes beyond watching a chatbot struggle. Job interviews are one of the few situations where people are expected to explain why they deserve to be chosen, often after navigating an application process that can feel impersonal and exhausting. Watching someone turn the tables on the interviewer, even an artificial one, offers a kind of satisfaction most can get behind.
And while some people are intentionally trying to break the technology, others have encountered AI interviewers that seem perfectly capable of doing that themselves.
In one viral video from 2025, an AI interviewer repeatedly uttered the phrase “vertical bar pilates” for nearly 30 seconds during what was supposed to be a job interview. Kendiana Colin, the college student who shared the encounter, said she wouldn’t participate in another AI interview.
“It was like she was going through a tongue twister,” Colin told Gazetteer SF. “It was really creepy.”
This Tweet is currently unavailable. It might be loading or has been removed.
For some job seekers, the issue isn’t even whether the chatbot works. It’s that they’re being asked to interview with one at all.
Research from hiring platform Greenhouse published earlier this year found that 38 percent of surveyed job seekers had walked away from a hiring process because it involved an AI interview. Another 21 percent said they hadn’t realized AI would be involved until the interview began.
This Tweet is currently unavailable. It might be loading or has been removed.
Employers, however, see opportunities that extend well beyond having a chatbot ask introductory questions.
Companies are making AI interviews part of the hiring process
Earlier this month, TechCrunch reported that technical hiring platform HackerRank was making its AI interviewer, Chakra, generally available after more than 500,000 interviews during beta testing. Unlike a chatbot that simply works through predetermined questions, Chakra can observe candidates completing technical assignments, ask follow-up questions, and evaluate how they approach problems.
For employers, that could mean assessing more applicants without requiring a human interviewer for every interaction. But other companies are using AI interviews to test something different: how well candidates can work with the technology themselves.
This Tweet is currently unavailable. It might be loading or has been removed.
Consulting firm McKinsey, for example, is experimenting with an AI-powered client simulation for some entry-level business analyst interviews, according to an Oct. 14 report from the Wall Street Journal. Candidates interact with a simulated client while human interviewers observe their performance. The firm is also reportedly testing how applicants use its internal AI assistant, Lilli, to work through business problems.
That creates an interesting distinction. Knowing how to use AI can be a qualification employers want, but using it without permission during an interview can still be considered cheating. According to a survey by the UK’s Institute of Student Employers, 66 percent of surveyed employers had identified or suspected unauthorized AI use during interviews.
The concern is understandable. Someone secretly relying on generated answers might appear more knowledgeable than they actually are. But there’s also an irony in companies embracing AI to evaluate candidates while investing in technology to catch applicants doing the same. Accenture, for example, says it uses AI to screen résumés, match candidates to jobs, and assist with assessments, while also employing proctoring tools and AI-detection methods to flag unauthorized AI use. HackerRank similarly monitors candidates’ screens for suspicious activity, even as it allows them to use its own AI assistant when employers enable it.
This Tweet is currently unavailable. It might be loading or has been removed.
When both sides use AI, who can you trust?
For applicants, the rules aren’t always obvious. In Greenhouse’s 2026 survey, 70 percent of respondents said they weren’t clearly informed upfront that AI would assess them.
And employers aren’t the only ones worried about being deceived. Research attributed to Gartner suggests only about half of job seekers believe the postings they’re applying to are legitimate. As AI makes it easier to generate convincing applications and automate interviews, both sides have more reasons to question who, or what, they’re dealing with.
This Tweet is currently unavailable. It might be loading or has been removed.
That could make something as ordinary as meeting face-to-face more appealing. Another Gartner finding suggests 62 percent of candidates would be more likely to apply for a job requiring an in-person interview. After years of moving more of the hiring process online, actually sitting down with another person could become a way to establish trust rather than just another step in the application process.
Companies may be able to automate more of hiring than ever before. But if neither side can be sure who’s actually on the other end of an interview, will getting hired eventually require going back to meeting face-to-face?