Fair Lending & Compliance

AI-based lending in the GDPR era

Zest AI team
July 23, 2018

The regulatory landscape for data handling changed substantially on May 25, 2018, when the EU’s General Data Protection Regulation, or GDPR, went into effect.

The goal of the GDPR is to protect personal data and promote fair and transparent personal data use practices in a world of increasing data collection. This new regulation broadly affects organizations, government agencies, and companies established within the EU, as well as companies outside the EU that collect, use, or share personal data of EU persons.

More businesses and governments are adopting artificial intelligence (AI) technologies such as advanced machine learning to make faster and better decisions. The GDPR requires financial firms adopting AI to meet a number of requirements when using it to automate credit decisions, including:

  • Providing transparency – Lenders must give applicants detailed disclosure about the use of automation to make credit decisions, the personal data used, the logic involved, and the potential for denial of their applications.
  • Honoring individuals’ rights – Applicants have a right to access or delete their personal data, restrict its use, and obtain a machine-readable copy of it.
  • Providing an appeals process – Applicants are entitled to an explanation of the credit decision and must be allowed to appeal automated credit decisions to a human decision-maker who considers additional evidence not previously considered.
  • Demonstrating compliance – Lenders must be able to demonstrate how they comply with GDPR in auditable records, including documentation of how automated credit decisions were made.
  • Ensuring quality control and integrity – Lenders must maintain mechanisms to resolve quality issues and errors, remove biases, and audit decisions for integrity and consistency.
  • Implementing adequate security measures Lenders must implement adequate technical and organizational security measures to protect the personal data they process.

These are no trivial tasks. Many of the tools available today to explain how AI makes decisions are flawed or hugely time-intensive. Lenders wanting to implement machine learning-based underwriting can drive meaningful revenue growth for their businesses, but can’t capture these gains without thorough explainability so they remain within the requirements of the GDPR era.

latest
March 25, 2024
Women’s History Month: sisters are doin' it for themselves
February 27, 2024
Zest Cares — we mean it when we say it!
February 26, 2024
Credit & Risk
Leaning downhill: rethinking risk in lending