
A job seeker used ChatGPT Voice to answer an AI recruiter's calls.
The two bots talked for 10 minutes, and no human followed up.
This reflects a broader trend where AI interviews are common, with 63 percent of job seekers reporting such encounters.
What happened
A job seeker named Christopher, after five unanswered AI interviews with recruiter 'Riley' from IT firm Everforth Apex Systems, fed his background into ChatGPT Voice and let it take a 10-minute call with Riley. The two bots exchanged pleasantries and got stuck in a loop about onboarding.
Why it matters
Christopher's experiment highlights a growing standoff where companies deploy voice AI to screen candidates, and applicants increasingly use AI to respond. A second 23-minute bot-to-bot call with a fake candidate named Don Dickner also ended with no human follow-up, leading Christopher to call it a 'slop flywheel.'
What to watch
A recruitment VP, Mark Monaghan, calls this the 'next logical stage' and says, 'If you're going to send a bot to me, I'll send a bot to you.' AI interview detection is already a market, with startups like Ribbon promising to flag 'overly scripted, AI-assisted, or coached' responses.
Ask the AI about this article →
Christopher's experience reflects how AI has seeped into every step of job hunting, both for applicants and recruiters. Recruiters, overwhelmed by applications, increasingly use voice AI to screen candidates, with 63 percent of job seekers reporting an AI interview, according to recruitment platform Greenhouse. Voice AI is tricky, as agents struggle with accents and interruptions, but the volume of applications has made it a practical tool.
At the same time, candidates have turned to AI to respond, leading to a kind of arms race where each side sends synthetic personas at the other. AI recruitment startups like Ribbon now offer to flag 'overly scripted, AI-assisted, or coached' answers, acknowledging the practice is widespread.
Christopher's story, where a ChatGPT persona and Riley exchanged pleasantries and looped on standard questions, suggests a future where bot-on-bot interviews may become routine. While some see it as duplicitous, Mark Monaghan of IQor frames it as a logical evolution. His comment, 'If you're going to send a bot to me, I'll send a bot to you,' captures the mutual suspicion and the potential for AI-mediated hiring to become a hollow exercise, as data flows from one synthetic persona to another with no human outcome.
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
ZeroDrift Inc. today introduced Guard for Agents, a service that converts written company policies into enforc…
Cloud AI capital expenditure continues to exceed market expectations, raising concerns about overheating and a…

Zhen Ding Technology showcased 800G, 1.6T, XPO, and NPO high-end optical transceiver modules, plus ultra-high-…

Nokia has opened a new research and development center in Riyadh, Saudi Arabia, focused on AI-powered network…

Anthropic updated the system prompt for Claude 5.1, adding a strict ban on reproducing song lyrics, poems, or…

President Trump sharply rejected the US backlash against AI data centers, writing on Truth Social that communi…
