Job recruiters are meeting an uncanny new kind of candidate: the AI avatar. These digital humans mirror an applicant's voice and likeness while being something else entirely, and they are turning up in first-round interviews. A Sydney-based recruiter, Lana Kersanava, says she has encountered them over the past six months. “At the beginning, it's like you feel like something is off,” she noted, adding that the rendering is remarkably realistic.
The Tell That Gives Them Away
The giveaway is not the visual presentation, which holds up under scrutiny. Instead, it is the answers themselves—delivered with a polished consistency that never wavers across a sequence of questions. After roughly five minutes, Kersanava ended the interview and disqualified the candidate. She has now done this three times. The pattern is telling: all three avatar candidates applied for English-speaking roles as non-native speakers, leaning on the technology to cover a gap in language proficiency. That is a different proposition from someone faking a work history.
This behavior lands in a hiring market with a documented bias problem. Research indicates that recruiters are up to 19% less likely to follow up with job seekers from immigrant and ethnic minority backgrounds than with equally qualified candidates. None of this makes an undisclosed avatar acceptable, but it does complicate the framing of these candidates as straightforward fraudsters.
The System They Are Responding To
Applicant tracking systems already use AI to filter CVs before a human reads them, and recruiters increasingly deploy AI avatars to conduct first-round interviews themselves. Candidates are now answering machines with machines. The resulting AI-on-AI war is rewriting the hiring playbook, pushing employers toward live technical walk-throughs, scenario challenges, and roleplay simulations that are harder to automate through. The volumes explain the arms race. Nearly two-thirds of candidates now use AI in applications, and 87% of US hiring managers use it in hiring, alongside a 239% surge in AI-generated applications.
The adoption of AI in hiring has been accelerating for years. From automated resume screening to chatbot pre-interviews, the recruitment process has become increasingly digitized. The pandemic-fueled shift to remote work accelerated this trend, as video interviews became the norm and software tools promised to save time and reduce bias. However, the same tools have also made it easier for candidates to game the system. AI-generated application materials, deepfake videos, and real-time voice modulation are no longer theoretical threats but practical realities. The rapid proliferation of generative AI tools has democratized access to sophisticated impersonation, allowing even non-technical users to create convincing avatars in a matter of hours.
How Common Is This?
A 2026 Greenhouse report found that 91% of US hiring managers have encountered or suspected AI-generated answers during online interviews. Across 19,368 live interviews analyzed between July 2025 and January 2026, 38.5% of candidates were flagged for AI-assisted behavior. The trajectory points one way. Gartner projects that by 2028 one in four job candidate profiles globally will be fake, spanning AI-generated profiles, deepfake video interviews, and fabricated work histories. Barriers to entry are minimal. Palo Alto Networks found that it takes as little as 70 minutes for someone with no image manipulation experience to build a fake candidate capable of passing a video interview.
The implications are staggering for the labor market. If a candidate can be fully virtual, what does that mean for trust, skill validation, and the very concept of a job interview? The interview has historically been a human-to-human exchange designed to assess not only technical competence but also cultural fit, communication style, and interpersonal rapport. When one party is a sophisticated algorithm, the dynamic collapses. Recruiters are forced into a game of cat and mouse, constantly updating their detection methods as AI-generated avatars become more advanced. Some companies are turning to specialized detection software that analyzes micro-expressions, eye movement, and vocal patterns, but these tools are far from foolproof and can produce false positives, penalizing nervous or neurodivergent candidates.
Employers Are Retreating to the Room
Seventy-two percent of recruiting leaders now run at least one in-person stage specifically to counter AI-assisted fraud, with Google, McKinsey, and Cisco reintroducing mandatory face-to-face rounds. That solves detection at the cost of everything remote hiring was meant to deliver. Candidates who cannot travel, including many of the international applicants this technology most tempts, lose access first. This retreat from remote hiring represents a significant reversal of the flexibility gains made during the pandemic. It also raises equity concerns: candidates with financial constraints, caregiving responsibilities, or disabilities may be disproportionately excluded from face-to-face rounds, reinforcing existing barriers to employment.
The in-person interview, while not immune to dishonesty, reduces the ability to deploy real-time AI assistants or deepfake avatars. But it is not a complete solution. Some companies are experimenting with identity verification systems that compare video interviews to government-issued IDs, while others require candidates to sign integrity pledges or undergo background checks that include a live technical test. Still, these measures add friction and cost to the hiring process, and they may not keep pace with the sophistication of AI generation tools. As Gartner's projection suggests, the war between deception and detection is likely to intensify, with no clear end in sight.
The Screening Was Never That Good
There is an awkward premise underneath the panic: that the automated filtering being defended actually works. The case against AI CV screening has been made repeatedly and forcefully, largely on the grounds that it optimizes for keyword matching rather than capability. Candidates optimizing against that filter with better tools is a predictable outcome rather than a moral collapse. Kersanava's five-minute detection rate also suggests that the current generation of avatars is not fooling attentive humans. What it is doing is making the first round worthless for everyone. If the machine screens the machine, the interview stops measuring anything, which is an argument for changing the process rather than policing the candidates.
The root cause of this problem may be the over-reliance on automated screening systems in the first place. These systems are designed to reduce recruiter workload, but they often do so by filtering out the very candidates who might thrive in a role. A 2020 study from the Harvard Business Review found that AI resume screening disproportionately penalizes gaps in employment history and non-traditional career paths, which are more common among women, caregivers, and immigrants. By optimizing for a narrow set of keywords, these systems narrow the talent pool and force candidates to adopt gaming strategies to get past the initial gate. The rise of AI avatars is a natural next step in this arms race: candidates are fighting fire with fire.
Some industry observers suggest that the solution lies in a hybrid approach that combines AI efficiency with human judgment. For example, AI could be used to flag inconsistencies in an interview or to screen for basic qualifications, but final decisions should rest with human recruiters who can read nuanced social cues and probe deeper into a candidate's experience. Another proposal is to shift away from traditional interviews altogether and rely on skills-based assessments, portfolio reviews, and standardized work samples that are harder to fake. Companies like IBM and Unilever have already piloted gamified assessments and asynchronous video interviews, but these too can be manipulated with AI tools if not carefully designed.
The broader context is a labor market that is increasingly mediated by technology, for better and worse. Job seekers are under immense pressure to stand out in a crowded field, and the tools available to them have never been more powerful. The ethical dilemma is not new—people have always embellished resumes and rehearsed interview answers—but the scale and realism of AI-generated deception are unprecedented. The question is whether hiring systems can evolve to keep pace, or whether the interview itself will become an obsolete ritual. As the data suggests, the current trajectory is unsustainable. The first round of interviews, if conducted through AI interfaces on both sides, risks becoming a meaningless exercise in mutual deception. Employers are therefore wise to reintroduce in-person stages, not as a punitive measure, but as a way to restore the human connection that has been lost in the automation of hiring. Yet the cost is real, and the candidates who lose out are often those who stood to benefit the most from remote and digital-first hiring processes.
The five-minute detection rate that caught Kersanava's attention is perhaps the most hopeful sign. It suggests that attentive humans can still spot the difference between genuine engagement and scripted perfection. The same may not hold true for AI detectors, which are vulnerable to adversarial attacks and false positives. Ultimately, the debate over AI avatars in job interviews is not about technology or fraud; it is about what we value in the hiring process. If we value efficiency above all, the machines will win. If we value authenticity and potential, we need to design processes that reward honesty and capability over polish and optimization. That will require a fundamental rethink of how we hire, but it may be the only way to preserve the integrity of the workplace.