Zest AI launches new run prediction model to reduce systemic bias in lending

Zest AI, a leader in artificial intelligence-based lending software, today announced the launch of Zest Race Predictor (ZRP). This open-source machine-learning algorithm estimates a person’s race/ethnicity using only their full name and home address as input.

ZRP can be used to analyze racial equality and outcomes in critical areas such as healthcare, financial services, criminal justice, or where it is necessary to assign race or ethnicity to a population dataset when data is absent race/ethnicity. The financial services industry, for example, has struggled for years to achieve fairer results amid allegations of discrimination in lending practices. Better judgment can help reverse this legacy of prejudice.

ZRP improves on the most widely used racial and ethnic proxy method, Bayesian Improved Surname Geocoding (BISG), developed by RAND Corporation in 2009. In several tests against BISG, ZRP was able to identify African Americans 25% more often, 35 % fewer African-Americans than non-African-Americans, and 60% fewer whites than non-whites.

A more accurate execution forecast will help the entire credit ecosystem:

  • Lenders will be better able to identify unfair outcomes to improve models.
  • Regulators will have a better tool to enforce fair lending rules that promote fair access to credit products that can help people of color get better credit scores.
  • Lenders will benefit from knowing that your race and ethnicity are more accurately reflected, along with your credit history.
Translate »