Is Our Money Safe with AI? A Deep Dive into Combating Financial Fraud

June 30, 2023

On a recent panel at a technology conference, two prominent figures in the world of financial technology and fraud prevention, Laura Spiekerman, co-founder and president of Alloy, and Trisha Kothari, co-founder and CEO of Unit21, engaged in an enlightening discussion about the impact of artificial intelligence.

Their conversation revolved around the implications of AI in combating financial fraud, the possibility of global regulations, and the challenges associated with innovation in AI models.

New call-to-action

Combating Financial Fraud with Trisha Kothari and Laura Spiekerman: Panel Recap

Laura Spiekerman initiated the conversation by addressing the distinction between financial crimes and fraud. She underscored that despite overlaps, these are separate issues that require different mitigation strategies. Spiekerman highlighted the use of AI as a powerful tool in fighting fraud by analyzing patterns and detecting anomalous behavior.

Adding her insights, Trisha Kothari mentioned the sophistication of modern fraudsters. She stressed the need for consistent data collection and shared how Unit21 is enhancing the data collection process. Kothari suggested that AI's strength lies in its capability to dissect vast datasets, identify patterns, and flag potential fraudulent activities. She emphasized the complexity of fraud and the constant innovation in schemes that keep even AI on its toes.

Both experts agreed on the necessity of proactive data analysis in fighting fraud. They considered the act of combating fraud as a perennial game of whack-a-mole, and the necessity for companies to stay ahead. The speakers also discussed ethical considerations in utilizing AI, such as discrimination and biases. They underlined the responsibility of companies to ensure that their AI systems don't inadvertently disadvantage or discriminate against any group.

Transitioning to regulatory aspects, Spiekerman questioned the feasibility and necessity of global regulations in combating financial fraud. The duo acknowledged that finding a balance between protecting consumers from scams and ensuring access to products is paramount. They also discussed the importance of explainability in AI models and the need to communicate these to auditors and regulators effectively.

Discussing ways to fight financial fraud, the speakers highlighted the significance of efficient data collection. They pointed out that banks' data is usually siloed, making it hard to correlate different data sources to identify fraudulent patterns. Spiekerman stressed the need to use a centralized system and adopt the newest technologies and AI models quickly.

Talking about investments, Kothari indicated that as long as fraud exists, there is room for companies like theirs in the market. The discussion also touched on the fear of AI replacing jobs, where Kothari opined that while some jobs may become obsolete, humans will always find new work due to their inventive and innovative nature.

Reflecting on the SVB downfall, Spiekerman mentioned how it revealed the shortcomings in onboarding processes and the potential for fraud when fast-tracking these processes. Kothari agreed, stressing the importance of robust core infrastructure for making sense of data to be useful in models.

Is Our Money Safe with AI Key Takeaways:

Summing up, the key takeaways from the discussion involve the significance of AI in combating financial fraud, the need for centralized and efficient data systems, the challenge of making AI models explainable, and the impact of regulatory issues.

The speakers underscored that as fraudsters evolve, the methods to detect and prevent fraud must also advance, highlighting the perpetual nature of this cat-and-mouse game.

  1. Distinction between financial crimes and fraud: Both are overlapping but different issues requiring distinct strategies. The speakers emphasized the importance of understanding the nature of these crimes for appropriate action.
  2. AI in fighting fraud: AI's power lies in its ability to analyze large data sets and spot patterns indicative of fraudulent activity. However, this requires regular data collection and adaptation to the evolving nature of fraudulent schemes.
  3. Ethical considerations: AI systems should not inadvertently disadvantage or discriminate against any group. The responsibility lies with the companies to ensure fairness in AI operations.
  4. Global regulations: While addressing the feasibility of global regulations, the balance between consumer protection and product access was stressed. The importance of making AI models explainable for effective communication with regulators was also discussed.
  5. Data collection and use: Efficient data collection is crucial in identifying fraudulent patterns. Current banking data is often siloed, complicating the process. A centralized system can significantly enhance fraud detection combined with the latest technologies.
  6. AI and jobs: While there's fear of AI replacing jobs, it was pointed out that humans, due to their innovative nature, will always find new work.
  7. Lessons from the SVB downfall: It highlighted the potential for fraud when fast-tracking onboarding processes, emphasizing the importance of robust core infrastructure to make data useful for models.

The discussions underline that as fraudulent methods evolve, the strategies to combat them must keep pace, emphasizing the ongoing nature of the fight against fraud.

Bad Fraud Practices Hand-Out

Getting started is easy

See first-hand how Unit21
can help bolster your risk & compliance operations