11,000 per minute
Every minute, 11,000 job applications are submitted on LinkedIn. That is 45 percent more than a year ago.
The mountain of job applications that no human reads anymore
Half of those applicants let AI do the work. ChatGPT writes the cover letter, tailors the CV to the job description, and sends it off. What comes out are piles of almost identical letters, in which a recruiter can no longer tell the difference by hand.
So the employer also takes a machine.
The recruiter is also getting a machine
52 percent of recruitment teams are deploying autonomous AI agents this year. 79 percent of all AI in recruitment is at the front end of the process: sourcing and screening. An agent reads a CV in less than 5 seconds, whereas a human takes 5 to 10 minutes. Some interviewing agents conduct 2,000 screenings a day.
A model is now sitting on both sides of the table. The candidate lets AI search and apply. The employer lets AI sort and select. And for the first time, the resulting shortlist has not been made by a human.
Here is where your AI visibility comes in.
An agent that matches candidates with employers does precisely what every language model does. They work with what they can read about you. If your story is clear, they weigh you up. If it is not there, you do not exist in their consideration.
Kim no longer applies for jobs, her agent does it
Kim won't be searching herself later. She'll let her agent do the searching. “Find employers in my field that suit someone like me, and apply where it's a match.” The agent reads, compares and selects three names to send. Will you be one of those three, or will it skip you?
He makes that choice based on what can be found. Your vacancies, your culture page, your reviews, the places where people talk about you. What is consistent, he remembers. What is unclear, he simplifies. And what is nowhere to be found, he does not take into account.
You filter by findability and are filtered on it yourself
On your side, the reverse happens. Your recruiter has an agent scour the market for candidates. That agent assesses people based on what it finds out about them. Anyone who doesn't have their work publicly readable gets filtered out, no matter how good they are.
That exact thing will happen to you soon as an employer, if your employer brand is unreadable to a model. You filter candidates on findability; you yourself are filtered on findability. The reference check has been turned around, and it is now running fully automatically.
The scarce candidate you never saw coming by
Think of a scarce role. A data engineer, a specialised nurse, a mechanic you've been looking for for months. That candidate's agent looks around, weighs up five employers, and you're either one of them, or you're not. If you're not among them, you don't get a rejection. You get nothing. You only realise you were competing once you've already lost.
A machine reads differently from a human
The numbers make a return to manual processing impossible. With 11,000 applications a minute, no recruiter reads every letter anymore. The machine becomes the first reader, on both sides of the process. And a machine reads differently from a human. Atmosphere escapes it. It has no gut feeling. It only sees what is in black and white, in language it can pick up.
The story that you no longer write yourself
I worked in recruitment for ten years. The whole time revolved around one thing: control over the narrative. What the job advertisement says, what the hiring manager says, what's on the careers site. That story had an owner, and that owner was you. Now, an agent reads everything floating around about you and makes its own single summary of it. That summary is your first impression, and you didn't write it.
Two types of employers and little room in-between
For such an agent, there are only two types of employers left. The employer with a clear, consistent story that he can retell, and who makes the list. And the employer he cannot read properly anywhere, and whom he skips for lack of guidance. Between those two, there is little room.
For employers, that means three things. What is still a manual selection today will in six months be an agent doing the pre-sorting. What still feels optional today will in a year be the condition for taking part. And what you have not made readable for a model today will later cost you the candidates you never saw pass by.
What you can do this week
Put the answers to the real questions on your own domain, in plain sentences that an agent can copy word for word: what you offer, how you treat people, where someone progresses with you. Make sure your story says the same thing everywhere, so a model doesn't have to guess which version is correct. And check for yourself what the models are saying about you today, because that is the fairest measurement of where you stand.
The reflex is to wait until those agents are actually here. They are already here. They are reading you now, building their impression now, and that impression will soon decide whether you make the selection.
Candidates no longer ask Google. They let their AI choose. We ensure that you are the answer.
My name is Bryan Peereboom, founder of AI Rebels. We make employers visible in language models.