by Rat Race Rebellion September 6, 2026
Subscribe to our daily newsletter to get the latest vetted remote job leads delivered straight to your inbox.The cost of applying to a job has dropped close to zero.
AI tools can customize a resume, generate a cover letter, and tailor application responses to a specific role in minutes. The constraint that once forced job seekers to be selective, time and effort per application, has largely disappeared. Application volume has followed. According to The New York Times, LinkedIn now processes roughly 11,000 applications per minute – up about 45% year over year. Most of that volume is not handcrafted.
Employers haven’t absorbed this passively. The majority now use some form of AI in their recruiting process, including resume screening and candidate ranking. Modern applicant tracking systems can generate shortlists before a human recruiter has read a single application. The automated screening layer isn’t making the same judgment a recruiter would make after reading a résumé. Depending on the system, it may be matching qualifications against the job description, ranking applicants by role criteria, or flagging materials for human review. The practical consequence is that getting into a recruiter’s consideration set can increasingly depend on how well an application survives the systems in front of it.
Greenhouse CEO Daniel Chait has described the result as an “AI doom loop”: job seekers use AI to apply to more jobs, employers use AI to filter them out, both sides escalate, and the process gets worse for everyone.
The signal problem
There’s a way to think about what’s happening that goes deeper than “the market is more competitive.”
Hiring is fundamentally about inference. Employers can’t directly observe whether a candidate is capable, genuinely interested, or worth the effort of an interview. They use application materials to infer those things. A well-crafted cover letter used to signal something – care, communication ability, genuine interest in this particular role. A resume customized to a specific company used to signal effort and fit. A thoughtful response to an application question used to suggest something about how a candidate thinks.
When these things become cheap to produce, they carry less information. When an AI-generated cover letter can look much like one a candidate spent an hour carefully writing, the finished document carries less information about how much effort, interest, or communication skill produced it. If every applicant can produce one in two minutes, a polished application no longer tells you much about the applicant.
This is what the “doom loop” is actually doing underneath the volume problem. It’s not just creating noise. It’s eroding the informational value of application materials for employers trying to identify strong candidates and for job seekers trying to demonstrate that they are one.
AI amplified this dynamic. But it didn’t create it from nothing. A softening white-collar hiring market, a more competitive remote job pool, fewer openings in certain sectors – all of these were already applying pressure. What AI has done is accelerate the pace at which certain signals have lost their usefulness.
None of this means job seekers should stop using AI. There’s nothing wrong with using AI to spend less time on formatting and structure leaving more attention for the substance that actually differentiates you. The distinction is between using AI to communicate evidence you actually have and using it to manufacture the appearance of evidence you don’t. AI can help you describe a project more clearly. It can’t make the project exist.
Did you find this interesting? Browse similar posts right here.What still means something
The signals that remain informative are the ones that are still expensive to produce.
A referral is the clearest example. Referrals make up a small share of total applicants but account for a disproportionate share of actual hires, a gap that reflects the power of a signal that can’t be mass-generated. A cold application asks an employer to infer your quality from materials that anyone can now produce in minutes. A referral adds something different: a person whose judgment the organization may already have reason to trust, providing information about a candidate that’s harder to fake and harder to scale. AI hasn’t cheapened that signal. It’s made it more valuable by comparison.
But a referral is the most visible example of a broader category. A portfolio of actual work – code that ships, writing with a public track record, demonstrated results in a specific domain – is harder to manufacture than a well-formatted résumé. A relationship built with a recruiter before a specific role is open is harder to replicate at scale. Domain expertise that’s visible and verifiable — someone who teaches what they know, writes specifically about an industry, or has built things that exist outside a job application — provides evidence that an AI-enhanced submission can’t substitute for.
Specificity is another signal that’s difficult to commoditize. AI can make almost any résumé sound polished. It can’t legitimately manufacture the particulars of work you’ve actually done: the problem you inherited, what you changed, who it affected, and what happened afterward. As generic professional language becomes easier to produce, concrete evidence becomes more valuable.
These aren’t new strategies. What has changed is the relative value of investing in them. The cold application funnel has gotten noisier; the signals that travel outside that funnel have become relatively stronger.
The Bottom Line
The tempting response to an application environment flooded with AI-generated volume is to generate more volume of your own – more applications, more customization, more throughput. That’s a rational individual response to the situation. It’s also what’s producing the “doom loop.”
The more durable question is different: what can you offer that isn’t easily replicated by the same tools everyone else is using?
That might be a referral from someone inside the company. It might be demonstrated expertise in a specific domain. It might be a relationship that puts you in front of an opportunity before it goes public. It might be a body of work that speaks for itself.
AI made it cheaper to apply. It also made the application, by itself, a weaker signal. The job seekers who stand out will increasingly be the ones who can provide evidence that’s harder to mass-produce – relationships, demonstrated expertise, specific results, and work that exists beyond the application itself.
Didn’t find what you were looking for? Check out these related roles and resources- Behind The Listings – What ‘Remote’ in a Job Listing Actually Tells You – and What It Doesn’t
- Hiring Company Roundups
- Company Guides – 10 Companies Hiring Remote AI Training Workers in 2026 (and How to Get Started)
- Side Gigs 2026
The post AI Made Applying Easier. It Also Made Standing Out Harder. appeared first on Rat Race Rebellion.
* This article was originally published here
