The job interview of the future might look like this: a virtual-reality headset drops you into a simulated workday. Or you play a video game that tests your decision-making in real time. Or an AI walks you through a problem-solving exam tailored to the exact role you're pursuing.
It sounds futuristic. But the reason companies are exploring these tools is grimly familiar — the way we hire people today simply doesn't work.
The System Is Already Failing
Ask any hiring manager or job seeker to describe the process, and "crapshoot" is the word that keeps coming up. A Robert Half survey of 2,200 U.S. hiring managers found that nearly a third had made a bad hire in just the past two years — most often because they'd misjudged a candidate's skills or cultural fit.
The problems begin before anyone sits down for an interview. Application tracking systems, the algorithms companies use to filter large volumes of candidates, routinely eliminate strong applicants early. And as job seekers increasingly use AI to polish their materials, employers aren't necessarily shortlisting the best people — just the best-presented ones. "Candidates' growing use of AI to craft applications means employers aren't necessarily shortlisting the best candidates, but those who present better," says Matthew Bidwell, a management professor at Wharton.
Then come the interviews themselves: multi-round, loosely structured, full of questions like "Where do you see yourself in five years?" Hiring decisions frequently hinge on gut instinct — which Bidwell calls "a terrible predictor." Research by Jason Dana, who joins Penn's behavioral and decision sciences program in July, goes further: unstructured interviews don't just fail to identify the best candidate — they can actually undercut whatever useful signal exists. "How would you like to know that the surgeon working on you was selected only because he's a hunter or something like that?" Dana asks.
Structure Helps. Games Might Help More.
One straightforward fix is to make interviews more structured — standardizing questions, scoring candidates on defined criteria, and reducing the role of intuition. It's catching on. Greenhouse, a hiring-software platform, reports that customers conducted 15 million structured interviews last year, up from roughly 500,000 in 2015. "The more we rely on structure and the skills we're assessing, the more effective the interview is at finding out who's good at the job," says Greenhouse CEO Daniel Chait.
But structure alone still relies on candidates telling you what they'd do. Games can show you.
Richard Landers, an organizational psychologist at the University of Minnesota, describes a simulation built for a software sales role. Candidates are placed in a scenario — they're pitching to a company's manager, CFO, and IT director, two of whom have raised serious objections. The challenge: close the deal within seven simulated days. Every step the candidate takes, from scheduling a client visit to managing pressure from their own sales manager, is tracked and scored across both hard and soft skills.
The design of these assessments matters enormously, Landers cautions. Research shows that men tend to have more experience with game-style interfaces than women — a gap that must be engineered out deliberately. "The games have to be engineered to avoid those kinds of problems," he says.
AI Interviews: Faster, Cheaper, Potentially Fairer
The next frontier is AI-powered interviews. Unlike the early-generation tools that filtered résumés by keyword, today's AI interview platforms can assess far richer information — and focus specifically on the skills that predict job performance, says Euan Cameron, CEO of candidate-assessment platform Willo.
LinkedIn is already testing this with an automated hiring agent for small businesses. Promising applicants receive a short AI-conducted interview that probes their skills before a human ever gets involved. The payoff is speed and scale: rather than a recruiter spending days interviewing every viable candidate, the AI screens for fit first. "AI can screen for top candidates based on objective questions faster than a human trying to determine whether each candidate has the capabilities," says LinkedIn chief product officer Hari Srinivasan.
There's another potential advantage that's harder to quantify: reduced bias. Human interviewers carry unconscious preferences into every conversation. AI, at least in principle, doesn't. "I do like the idea that AI at least filters out some of the personal biases that a lot of interviewers will bring with them," says Landers.
VR: The Long Game
Further down the road, virtual reality could take immersive assessment to its logical conclusion. Landers points to a scenario where a hospital hiring triage nurses has candidates put on headsets and step into a simulated disaster site. Do they notice the right things? Do they take the right first steps? Every decision, or non-decision, is observable and scorable.
The barrier right now is cost. VR hardware remains expensive enough that widespread adoption is still years away. But the underlying logic is compelling: the gap between what candidates say they'd do and what they'd actually do is where most hiring mistakes live. "Maybe the extra levels of immersion and fidelity can really get something out of understanding what that person would really do," Landers says, "not just what they say they would do."
That's ultimately what every hiring tool is reaching for — and why, despite decades of failed promises from technology, the search for something better continues.