Sign in Subscribe
Financial services

AI in Financial Institutes: Risks and Regulatory Response

Strategic Insights

— 1 min read

As artificial intelligence (AI) becomes increasingly integrated into financial institute business operations, the Government Accountability Office (GAO) has sounded the alarm on the potential risks to data privacy and fairness, including bias in lending decisions and threats to cybersecurity.

The GAO report highlights the double-edged nature of AI, which can both improve customer service and detect threats, but also amplify existing risks such as harm to fair lending practices, conflicts of interest, and false information. The authors found that AI can perpetuate or increase the risk of bias in lending decisions, leading to credit denials or higher-priced credit for borrowers, including those in protected classes.

Key Takeaways:

  • AI can amplify existing risks in financial institute business operations, including harm to fair lending practices, conflicts of interest, and false information.
  • AI can perpetuate or increase the risk of bias in lending decisions, leading to credit denials or higher-priced credit for borrowers, including those in protected classes.
  • Poor-quality data can contribute to biased AI models, which can then reinforce discriminatory decisions based on those biases.
  • Factors that increase risk include complex and dynamic AI models, poor-quality data, and reliance on third parties.
  • AI models deployed in financial institutions' environments face risks of producing misleading information, hallucinating entirely made-up answers, and exposing sensitive consumer data.

The GAO report also highlighted the need for new risk management guidance from regulatory bodies, particularly in areas where financial institution regulators lack the authority to investigate technology services.

Statistics:

  • 71% of consumers are concerned about the use of AI in the financial industry (Source: Nextgov).
  • 60% of institutions have experienced some form of AI-related issue in the past two years (Source: Nextgov).
  • The National Credit Union Administration lacks the authority to investigate financial institutions' technology services (Source: GAO).
  • Some regulators have issued AI-specific guidance, such as on AI use in lending (Source: GAO).

Sources:

  • Government Accountability Office (GAO) report: "Artificial Intelligence: Emerging Risks and Opportunities in Financial Institute Operations" (2023).
  • Nextgov: "Forget Dystopian Scenarios: AI is Pervasive Today and Risks are Often Hidden" (November 2023).
  • GAO: "Artificial Intelligence: Emerging Risks and Opportunities in Financial Institute Operations" (2023).

Read more