Why Your Job Hunt Feels Like A Lottery Thanks To AI
When Eric Thompson lost his job last year, he began a familiar but frustrating routine. He refined his resume, scoured job boards, and sent out applications, only to apply for countless roles that seemed to be fake. Thompson's frustration with these "ghost jobs" has now motivated him to advocate for federal legislation to ban them.
Thompson's experience is becoming increasingly common, highlighting one of the significant ways artificial intelligence is reshaping the job search. AI is turning what was once a manual, labor-intensive process into a high-volume, yet increasingly impersonal, numbers game.
The New Impersonal Numbers Game
The impact is visible across the board. Graduates in software engineering and computer science, once considered guaranteed hires, now face difficulties finding work. Early-career professionals in occupations heavily exposed to AI are seeing their employment prospects diminish. This widespread sentiment was captured by writer Kelsey Piper, who described the modern job hunt as feeling more and more like a lottery.
While this feeling resonates with millions, the underlying economics reveal a more nuanced situation.
An Economic Look At The AI Job Market
A key concept in labor economics, the Diamond-Mortensen-Pissarides framework, helps explain how unemployed individuals find jobs. The model is based on the idea that job hunting is a matching process filled with frictions, information imbalances, and search costs for both job seekers and employers.
Before the advent of large language models (LLMs) like ChatGPT, employers used advanced applicant tracking systems to screen resumes automatically. This process was efficient for them, but for workers, it was a tedious, manual task of tailoring resumes and repeatedly entering the same information into different portals. This, too, felt like a lottery where the effort was high and the payoff uncertain.
The release of ChatGPT dramatically altered this dynamic by empowering job seekers to submit applications en masse, leading to several key effects on the labor market.
The Unintended Consequences Of AI In Hiring
First, AI has made it more difficult to distinguish between genuine hiring efforts and serious job seekers. AI enables workers to apply for numerous jobs effortlessly, and it allows employers to post openings with equal ease. This surge in low-commitment employers likely explains the rise of ghost jobs. For an applicant, there is little difference between a ghost job and a posting from a low-commitment employer—both lead to applications vanishing without a trace.
AI has also made it harder for candidates to stand out. To differentiate themselves, job seekers will need to rely more on harder-to-game qualifications like strong portfolio projects, robust professional networks, and personal referrals.
This could lead to a divided job market. One segment would be high-volume and algorithmic, managed by LLMs, while the other would be relationship-based and mediated by humans, each with its own wage and employment trends. This model may explain the declining job prospects for early-career workers in AI-exposed fields. Recent graduates often lack the extensive work samples, professional networks, and proven track records that more experienced candidates use to their advantage.
More Noise Less Signal The Data On AI Hiring
One of the few academic studies on this topic, "Generative AI and Labor Market Matching Efficiency," by economists Emma Wiles and John J. Horton, reveals that the dynamics are still evolving. Their research found that employers using AI were 19 percent more likely to post a job and spent 44 percent less time writing the description. However, the resulting job posts were more generic and less informative for job seekers.
Crucially, the authors found no significant increase in successful matches, meaning AI has not improved hiring efficiency. Instead, it has introduced more noise into the system. They attribute this to "marginal jobs being posted by employers with lower hiring intent"—in other words, the market is being flooded by low-commitment employers.
A Path Forward Beyond Banning Ghost Jobs
Eric Thompson's proposal to ban ghost jobs targets a symptom rather than the root cause of the problem. Instead of attempting to return to a pre-AI reality, policymakers should focus on enhancing market transparency. This involves creating new frameworks that help both workers and employers signal genuine intent.
Some companies already provide hiring rates and timelines, but third-party platforms could standardize this by aggregating verified hiring data. While this groundwork is beginning, establishing it as a standard practice will require time. The feeling of playing a lottery in the job market is not a given; it is the outcome of a matching system that has yet to fully adapt to a powerful new technology.
