The hiring signal AI can’t fake
Trace hiring
Applying for a job used to take effort, and effort was a signal. AI removed it. Trace hiring adds back a signal AI cannot fake, by watching how a candidate thinks through a real problem. Here is what it is, and how to run it in minutes.
The definition
What trace hiring is
Trace hiring is a way to evaluate candidates by the traces they leave while working through a real problem, instead of the claims on an application. You watch how a person thinks, adapts, and builds. That is the part AI cannot fake for them.
The term was coined by Marvin Starominski-Uehara, a professor at Temple University, in an August 2026 piece for the World Economic Forum. He drew the idea from how termites coordinate complex work with no central direction, simply by leaving traces for each other to follow. The same logic runs Wikipedia edits and GitHub commits. Applied to hiring, it means designing a step that leaves an honest trace of how someone actually works.
Why now
The resume can’t carry the signal anymore
Applying for a job used to take real effort, and that effort quietly filtered out casual applicants. Generative AI removed the effort, and the flood arrived. The application still shows up. It just no longer proves anything.
More applications on LinkedIn in late 2024, while postings there fell 10.6%.
Of recruiters now spend up to half their week filtering AI spam.
Of hiring managers have caught candidates being dishonest.
Of the time an AI resume screener favored the white-associated name.
Figures compiled in the World Economic Forum’s 2026 piece on trace hiring, drawing on 2024 to 2025 recruiting data and surveys.
So every hiring team now faces the same question. How do you find genuine talent when the application itself no longer proves effort?
Why the obvious fix falls short
Automating the screen scales the bias
The common response is an AI applicant tracking system to handle the volume. Adoption nearly doubled in a single year. But these tools learn from past hiring, so they surface people who resemble whoever was hired before, and they carry the old patterns forward with them.
One study of three language models across more than 500 resume comparisons favored the white-associated name by nearly ten to one. Screening faster does not fix the signal. It just repeats the same mistakes at scale, with no human accountable for any of them.
The framework
The three stages of trace hiring
Stage 1
Respond to a real problem
Candidates answer a real question from the role, by video or in conversation. The ones leaning entirely on AI produce generic, interchangeable answers. The ones who engage produce something that is unmistakably them.
Stage 2
Work a live problem together
A shortlist works in small groups on a new problem, given on the day, with a tight deadline. Laptops and AI tools are allowed. Evaluators watch the process, not the output, noting who drives, who adapts when an idea fails, and who pulls a quiet teammate back in.
Stage 3
Reflect
Finalists answer one question. What did you learn from that exercise? Most describe the task or their team. A smaller number describe where their own thinking surprised them, and that kind of honest self-assessment is very hard to fake.
How to run Stage 1 in minutes
Stage 1 is where the value is, and it is the easiest to run at scale
You send a short challenge, a real question for the role. Candidates respond by video in a few minutes from their phone. You watch how they think. It sits after the resume and before the first call, so you walk into that call already knowing who is real.
That is what we built McCoy IQ to do.
- A real question for your role, not a quiz or a keyword filter.
- Every candidate gets the same challenge, so the field is level.
- AI helps candidates prepare their answer. It never scores or ranks them.
- Your team reviews the evidence and makes the call. No AI screening, no black box.
The instinct is to add friction back, longer applications, more hoops. That filters out good people too. The signal you want is effortless for a real candidate and useless for a bot.
Free to start. No credit card, no ATS integration required.
The real signal
The qualities that matter most
A resume won’t show you any of this. One to five minutes of someone talking through a real problem will. And you can watch it before you spend an hour on a call.
The shift
Watch what people do
Companies spent decades treating credentials, brand-name employers, and years of experience as proxies for quality. Those signals are now actively misleading. When one researcher analyzed more than 80 workplace studies spanning 60 years, he found that prior work experience has almost no link to how someone performs in a new role, and AI is thinning that link further.
What someone can do when the problem is real and the clock is running matters more than what they claim to have done alone.
The teams that hire well this decade will be the ones that watch what a candidate actually does, not the ones that screen the most applications the fastest. AI did not create this problem. It accelerated a reckoning that was already overdue.
Run trace hiring on your next role
Send one real question. See who is genuinely a fit before you spend a single call. Set it up in a few minutes.
FAQ
Common questions about trace hiring
What is trace hiring?
Trace hiring is a way to evaluate candidates by the traces they leave while working through a real problem, instead of the claims on an application. You watch how a person thinks, adapts, and builds, because that is the part AI cannot fake for them.
Who coined the term trace hiring?
Marvin Starominski-Uehara, a professor at Temple University, coined it in an August 2026 piece for the World Economic Forum, drawing on how termites coordinate work by leaving traces for each other to follow.
What are the three stages of trace hiring?
Stage 1, candidates respond to a real problem from the role by video or conversation. Stage 2, a shortlist works a new problem live in small groups while evaluators watch the process. Stage 3, finalists answer one reflection question about what they learned.
How is trace hiring different from an ATS?
An applicant tracking system scores or ranks people based on the words in their application, which AI can now write for them. Trace hiring ignores the claims and looks at how a candidate actually thinks through a real problem, and it keeps the decision with a human rather than an algorithm.
Is trace hiring just a video interview?
No. A video interview is scheduled, live, and often scripted. Trace hiring Stage 1 is a short async response to a real problem from the role, done on the candidate’s own time, so what you see is how they think, not how well they perform on a call.
How do I run trace hiring?
McCoy IQ operationalizes Stage 1. You send a short challenge with a real question for the role, candidates respond by video in a few minutes, and your team reviews how they think before the first call. It is free to start and needs no ATS integration.