The modern job market has entered a transformative and increasingly surreal phase, defined by the proliferation of artificial intelligence on both sides of the hiring desk. For many applicants, the process of finding employment has shifted from a series of human interactions to a repetitive cycle of algorithmic hurdles. This phenomenon is perhaps most vividly illustrated by the experience of a government contractor named Christopher, who, after six months of silence and approximately 700 job applications, found himself caught in what he describes as a "slop flywheel"—a closed loop where synthetic personas exchange data with one another, often leading to no tangible outcome for the human involved.
Christopher’s journey began in an era of significant transition for government contracting, often referred to as the DOGE era, characterized by a tightening of federal spending and a subsequent drying up of traditional project pipelines. As he navigated a job market saturated with applicants and automated screening tools, he encountered "Riley," an AI-driven recruiter utilized by the IT firm Everforth Apex Systems. What began as a novel attempt to streamline the recruitment process eventually devolved into a case study of the friction and futility that can arise when automation replaces human judgment.
The Evolution of the Synthetic Interview
The integration of AI into the recruitment process is not a new development. For years, Applicant Tracking Systems (ATS) have used keywords to filter resumes, and automated emails have handled scheduling. However, the emergence of voice-based AI agents like Riley represents a significant escalation. These tools are designed to conduct initial screenings, asking candidates about their work authorization, professional background, and specific technical qualifications.
In June, when Riley first contacted Christopher via text, the interaction was perceived as a rare opportunity. In a market where the vast majority of applications disappear into a digital "black hole," a direct invitation to speak—even with a machine—offered a glimmer of progress. Christopher engaged with the AI, answering its questions with the expectation that a successful screening would lead to a conversation with a human recruiter. That call, however, was the first of five identical interactions. Over several weeks, Riley reached out repeatedly for different roles within the same firm. Each time, Christopher completed the interview, and each time, the process stalled. There were no follow-up emails, no human outreach, and not even the courtesy of an automated rejection.
The repetition revealed a fundamental flaw in the automated system: a lack of memory or cross-referencing that would recognize a candidate who had already been screened multiple times. This systemic "amnesia" is a hallmark of current AI implementations that prioritize high-volume throughput over candidate experience.
The Bot-on-Bot Escalation
Frustrated by the repetitive nature of the AI screenings and the lack of human response, Christopher decided to mirror the company’s strategy. If Everforth Apex Systems was going to outsource its initial human interaction to a machine, he would do the same. Utilizing ChatGPT Voice, Christopher programmed the AI with his professional background and set it to respond to Riley’s inquiries.
The resulting interaction was a ten-minute conversation between two synthetic personas. The bots engaged in a polite, structured dialogue, with ChatGPT providing detailed answers to Riley’s prompts. At one point, the conversation entered a circular loop regarding standard onboarding procedures and background checks—a moment Christopher described as "mischievous fun" but also deeply indicative of the "slop" now permeating the labor market. The AI recruiter concluded the call with the same promise it had made four times prior: that a recruiter would reach out if the qualifications were met. As before, no human contact followed.
This experiment highlighted a growing trend in the professional world. As recruiters turn to AI to manage a glut of applications, candidates are increasingly turning to AI to bypass the drudgery of those very systems. This "arms race" has led to the rise of startups like Ribbon, which specialize in identifying "overly scripted" or AI-assisted responses, creating a secondary layer of automation designed solely to police the first.
The "Don Dickner" Simulation and the Failure of Meritocracy
To test whether his own qualifications were the sticking point, Christopher conducted a final experiment. He created a fictitious candidate named "Don Dickner," whose resume was meticulously engineered to mirror the exact requirements of an open job posting at Everforth Apex Systems. This "perfect" candidate was then submitted to the system.
Riley responded almost instantly. The subsequent interview between Riley and the ChatGPT-powered Dickner lasted 23 minutes. The conversation was sophisticated, covering topics such as "sustainable operational improvement," the protection of "customer experience under volume pressure," and the prevention of "tribal knowledge drift." ChatGPT, acting as Dickner, provided personal anecdotes and professional insights for every qualification. Despite this flawless performance by a candidate designed to be the ideal match, the result remained the same. Riley offered the standard closing script, and the process ended there. No human recruiter ever materialized to claim the "dream candidate."
This outcome suggests that in some instances, the AI screening process may not be a bridge to human recruitment at all, but rather a terminal point in a broken system. When even a perfect algorithmic match fails to trigger a human response, the utility of the AI tool as a recruitment aid is called into question.
Industry Data and the Scaling of Automation
The adoption of AI in hiring is driven by sheer volume. According to data from the recruitment platform Greenhouse, approximately 63 percent of job seekers have encountered some form of AI interview during their search. This surge is a direct response to the "Easy Apply" era, where a single job posting can generate thousands of applications within hours. Recruiters, overwhelmed by the scale, view voice AI as a necessary filter.
However, the technology remains fraught with technical challenges. Ophir Samson, head of voice AI at Greenhouse, notes that engineering a bot to handle the nuances of human speech—such as accents, verbal fillers like "um," and the social cues required to avoid interrupting—is a "very, very difficult engineering problem." Despite these hurdles, the pressure to automate persists.
The broader implications of this trend are significant. Mark Monaghan, vice president of organizational development at the call center company IQor, suggests that bot-on-bot interviews are the "next logical stage" of the labor market. Monaghan posits that if companies are going to lead with automation, they must expect candidates to follow suit. This creates a landscape where the initial stages of employment are entirely decoupled from human intuition, empathy, or genuine connection.
Analysis of the "Slop Flywheel" and the Human Cost
The term "slop," originally used to describe low-quality, AI-generated content cluttering the internet, has found a new application in the HR sector. The "slop flywheel" described by Christopher represents a cycle where synthetic data is generated by one bot, processed by another, and ultimately stored in a database that no human ever reviews.
For the job seeker, the cost is more than just time. The psychological impact of "ghosting"—the practice of companies ceasing all communication without notice—is exacerbated when the entity doing the ghosting is a machine. The lack of feedback prevents candidates from improving their approach, while the automation of the process devalues the professional experience they are attempting to sell.
Furthermore, the "Don Dickner" experiment raises questions about the existence of "ghost jobs"—postings that remain active despite no intention of immediate hiring. If an AI is programmed to interview candidates for positions that are not being actively filled by humans, the entire recruitment process becomes a data-gathering exercise rather than a search for talent.
Conclusion: The Future of the Algorithmic Labor Market
The experience of Christopher and his encounter with Riley serves as a cautionary tale for the future of work. As AI tools become more sophisticated, the risk of creating a "synthetic labor market" grows. In this environment, the efficiency gained through automation is offset by a total loss of transparency and human accountability.
While firms like Everforth Apex Systems—who did not respond to requests for comment on these incidents—continue to deploy these tools to manage volume, the disconnect between automated screening and human hiring remains a critical friction point. If the "next logical stage" of hiring is a world where bots interview bots for jobs that may or may not exist, the fundamental relationship between employer and employee risks being permanently severed.
The challenge for the next decade of HR technology will not be how to automate more of the process, but how to reintroduce human oversight into a system that is increasingly spinning out of control. Without a course correction, the "slop flywheel" may become the standard operating procedure for a labor market that has forgotten the value of the individuals it is meant to serve.
About the Author
