AI tools are increasingly replacing human judgment in settings such as resume screening, where they improve efficiency and decision consistency. Scholars warn that if an entire industry adopts the same algorithm, an “algorithmic monoculture” could cause systematic exclusion—candidates rejected by one firm’s algorithm would likely be rejected by all. MIT researchers examined these concerns systematically and found many arguments unconvincing. Using several models they showed that monoculture creates informational echo chambers that suppress exploration, making it harder for the best candidates to be discovered. However, bundling multiple hiring algorithms into an “ensemble algorithm” can mitigate this effect and, in some scenarios, outperform a polyculture of diverse algorithms. The study notes that if a single algorithm is substantially more accurate than the many used by different firms, monoculture may actually be preferable. Monoculture can also increase candidates’ bargaining power because firms compete for the same talent pool, potentially driving up wages. Agency‑related objections—such as candidates being unable to revise their resumes—lose force if resubmission is allowed. While a uniform algorithm might enable candidates to game the system by adopting a favored resume format, this advantage is not clearly greater than targeting a few specific algorithms. The authors invoke the “wisdom of crowds” principle, arguing that diverse independent decision‑makers typically identify higher‑quality hires than a single decision‑maker. Homogenized information can limit the discovery of novel talent, a particular concern for fields like science, art, or writing that rely on innovation. Introducing randomness or building ensemble algorithms are suggested ways to boost exploration. Ultimately, the benefits and drawbacks of algorithmic monoculture depend heavily on context and algorithmic accuracy, and further empirical work is needed.
Review