Building New Technology at FICO: Scott Zoldi
Scott Zoldi, Chief Analytics Officer at FICO, has a crucial advice for those using artificial intelligence (AI): Don’t get too comfortable.
“I’ve been in this industry and building AI models for a long time — 25 years or more — and about seven years ago, I got really concerned that people are getting too comfortable with AI,” said Zoldi. “Twenty-five years ago, you had to go through quite a bit of discussion to get anyone to adopt a machine-learning model. Today, it seems many people don’t even ask the basic questions,” he noted. “Mistakes are being made.”
A significant issue is that many new adopters of AI, drawn in by the current hype, are using models without the capability to explain how they work. This creates a clear risk of introducing or perpetuating biases that can affect customers, Zoldi warns.
Zoldi, who holds a doctorate in theoretical and computational physics from Duke University, submitted two patents in the past year to address these concerns. The first of these new patents, which has been granted, codifies AI model development in a blockchain to create an immutable record of who performed, tested, and verified the work.
“It’s not just a checklist,” he said. “It shows when mistakes are made, when things were rejected, and when things were remediated… it provides transparency, which is critically important.”
The second patent, which is pending, focuses on transparency as well—ensuring that the person who built the model documents how it was done. This way, they (or someone else) can look at it years later and understand what to monitor about it.
To encourage future generations of experts, Zoldi was the executive sponsor of the first FICO Educational Analytics Challenge, which collaborates with several historically Black colleges and universities to train students to work with complicated data. The first year’s focus was on identifying and mitigating bias in data collected by the Consumer Financial Protection Bureau. The upcoming year’s focus will likely be on payment fraud data, he said.
Key Takeaways:
- Don’t get too comfortable with AI: continuous scrutiny and understanding are crucial.
- Transparency is essential: blockchain and thorough documentation help in monitoring AI models.
- Encouraging future generations: initiatives like the FICO Educational Analytics Challenge train students to handle complex data responsibly.
By paying attention to these principles, the AI community can avoid perpetuating biases and ensure that AI technologies are both reliable and fair.