Technology
How Algorithmic Bias Reinforces Socioeconomic Inequality
Quick fact
A 2019 study found that a widely used healthcare algorithm was less likely to refer Black patients to extra care programs than equally sick white patients, because it used healthcare costs as a proxy for illness—and Black patients historically had lower healthcare spending due to barriers to access.
Why this is interesting
You might trust an algorithm to be fair because it's just math. But what if that math is quietly making inequality worse?
Read the full explanation
Understanding How Algorithmic Bias Reinforces Socioeconomic Inequality
Imagine you're a judge trying to decide who gets a loan. You look at the applicant's past financial history. If you only consider that history, you might reject someone who has had a few late payments years ago, even if they've changed. Now imagine an algorithm that scans thousands of such histories to find patterns. It might learn that people from a certain neighborhood (which historically was discriminated against) are more likely to default. It doesn't 'know' about race, but the neighborhood is a proxy. The algorithm then denies loans to people from that area, which makes it even harder for them to build wealth, and the cycle continues.
A deeper explanation
The core mechanism is a feedback loop: biased data → biased algorithm → biased decisions → further inequality → more biased data. Algorithms are trained on historical data that contains societal biases, such as discrimination in lending, hiring, or policing. They learn these patterns and apply them at scale. For example, predictive policing algorithms trained on past arrest data may direct more police to minority neighborhoods, leading to more arrests, which then reinforces the algorithm's belief that those areas are high-crime—even if the initial arrests were biased. This creates a self-fulfilling prophecy. The bias isn't just a one-time unfair decision; it accumulates, widening the gap between socioeconomic groups. Moreover, because these systems are often opaque, it's hard to challenge their decisions, making the loop even more difficult to break.