Fintech Week in Review – April | Perkins Coie

Weekly Fintech Focus

• Federal financial regulators issue an RFI to better understand financial institution usage of AI models and its implications on current regulations.
• CFPB rescinds policy statements that provided regulatory flexibility in response to pandemic challenges.
• National Futures Association issues regulatory outsourcing guidance.
• Lina Kahn announced as nominee for FTC commissioner position.
• FinCEN Director meets with a number of groups to discuss SAR filing.
• Prepared Remarks of FinCEN Director delivered at the Florida International Bankers Association AML Compliance Conference
• CFPB faced many more consumer complaints in 2020 than in the past.
• CSBS working with a number of states to develop a single fintech exam.
• An earned wage access company settled allegations of its services causing bank overdraft and insufficient fund fees.
• The UK FCA issues a feedback statement on open finance.

Federal Financial Regulators Seek Information on the Use of AI by Financial Institutions

On March 29, 2021, five federal financial regulatory agencies (the Federal Reserve, CFPB, FDIC, OCC, and NCUA) announced a request for information (RFI) to obtain input from financial institutions, trade associations, consumer groups, and other stakeholders about the use of artificial intelligence (AI) by financial institutions. The regulatory agencies intend to use the information received in the RFI to determine whether any clarifications from the agencies would be helpful for the use of AI by financial institutions in a safe and sound manner. The agencies identify laws and regulations that they understand could relate to financial institutions’ use of AI, including Section 39 of the Federal Deposit Insurance Act, the Gramm-Leach-Bliley Act, the Fair Credit Reporting Act, the Equal Credit Opportunity Act, the Fair Housing Act, and the prohibitions against UDAP/UDAAPs. The RFI also includes a list of supervisory guidance and statements and exam manuals that relate to financial institutions’ use of AI.

The RFI includes questions related to:

• Explainability – The agencies seek information on the strategies and challenges that exist for financial institutions related to explaining how AI models make decisions.
• Risks from Broader or More Intensive Data Processing Usage – The agencies seek information about how financial institutions address data quality and data processing for AI models.
• Overfitting – The agencies seek information about how financial institutions address the risks of overfitting in AI models – the risks associated with an algorithm learning from patterns in training data that are not representative of the population as a whole.
• Cybersecurity Risk – The agencies seek information related to AI vulnerabilities to cybersecurity threats.
• Dynamic Updating – The agencies seek information about how financial institutions manage risks related to dynamic updating – where the AI model is able to learn and evolve over time as it collects new training data. The risks arise when the results drift from the model’s past behavior or intended use.
• AI Use by Community Institutions – The agencies seek information about how community institutions may rely on other third-party AI service providers and the attendant risk and challenges.
• Oversight of Third Parties – The agencies seek information about the challenges and risks associated with financial institutions overseeing AI models developed or provided by third parties, and whether current agency guidance needs to be updated.
• Fair Lending – The agencies seek information about how financial institutions evaluate AI-based credit determination approaches to comply with fair lending laws, including addressing bias and complying with the Equal Credit Opportunity Act’s adverse action disclosure requirements.

CFPB Rescinds Certain COVID-19 Regulatory Flexibility Guidance

Effective April 1, 2021, the Consumer Financial Protection Bureau (CFPB)…