When AI makes every cover letter look polished, employers switch to signals that are harder to fake: past work and reputation.

This explores what employers rely on once AI makes every cover letter look good. The best evidence comes from one natural experiment. When Freelancer.com launched an AI tool that wrote tailored proposals, the link between a letter's fit to the job and whether the applicant got a callback fell by 51% Does AI cover letter writing change what employers value?. Letter quality also became much weaker at predicting interviews and job offers Does AI-generated cover letter access weaken hiring signals?. Employers didn't stop hiring. They switched signals and leaned on things that are hard to fake in one afternoon: prior work history, completed jobs and platform reputation.

The reason letters lost their value is the useful part. A cover letter never worked because of what it said. It worked because writing a good one cost effort, and that cost was lower for able workers, so effort stood in for ability. With AI tools that link breaks. Effort and signal quality now move in opposite directions, and polished proposals no longer predict whether the job gets done Why did AI tools break the effort signal in hiring?. Employers also became less willing to pay extra for workers whose proposals looked strong. Any replacement signal has to rebuild that missing cost. Kessler proposes two kinds Can hiring signals survive when AI makes cover letters worthless?. One is third-party vouching, like recommendation letters, where someone else puts their reputation on the line; this already shows measurable hiring benefits. The other is scarce time, like meeting in person or networking, which shows real interest because you can't do it for 200 jobs at once. That second idea is still untested.

Overall hiring rates stayed stable, but who gets hired changed. A simulation of the same market without written signals finds top-quintile workers hired 19% less often and bottom-quintile workers 14% more often Does cheap writing weaken hiring based on worker ability?. The market still runs, but it's less fair to the most capable people. There's also a signal employers can't see. Within the same worker, more time spent editing the AI draft goes with more hiring success, even though most people barely revise Does editing time on AI drafts predict hiring success?. Effort still matters, but it no longer shows in the finished letter. This matches a broader workplace pattern: people protect cues tied to identity, like their voice and where the work came from, while effort, attention and uncertainty disappear into the polished output Which workplace cues survive AI mediation and which disappear?.

Outside freelance platforms the picture is noisier. Greenhouse describes a 'doom loop': 41% of job seekers use AI prompt injections to get past screening filters, and 34% of recruiters spend half their week filtering spam. The survey backs up each part of the loop but doesn't show which side started it Are job applicants and employers locked in an escalating AI arms race?. Some new signals have the same weakness letters had. In a study with 1,725 recruiters, listing AI skills raised interview invitations by 8 to 15 percentage points, but certificates added only a little over simply claiming the skill Do AI skills help candidates get more job interviews?. The lesson across these notes is that a signal survives only if it stays costly or someone else vouches for it. A cheap self-description loses value as soon as AI can produce it.