EP 141: How To Understand and Fix Biased AI

Finance Practices

The financial sector has long been at the forefront of addressing AI bias, leveraging established practices to ensure fairness in algorithmic decision-making. Nick Schmidt highlights that big banks have been collaborating with consultants for over 25 years to develop fair algorithms, setting a precedent for other industries 1. Despite this, tech companies often overlook these established solutions, opting to innovate independently, which can lead to biased outcomes. Schmidt notes that using AI to fix AI is not only possible but can result in faster and fairer model development 2.

   

Regulatory Role

Regulations play a crucial role in promoting AI fairness across various sectors. Nick Schmidt suggests that impending regulations will likely push companies towards adopting practices similar to those in the financial sector, which have been effective in mitigating bias 1. He explains that while algorithms can initially reflect human biases, they have the potential to evolve and eliminate these biases over time, making AI systems fairer 3.