Artificial Intelligence, Affectionately Known as AI: Takeaways From 2023 AFSA Annual Meeting

Shareholders Sara Costanzo and Milos Gvozdenovic recently attended the American Financial Services Association (AFSA) 2023 Annual Meeting. At this event, attendees had the opportunity to connect with various keynote speakers, engage in educational breakout sessions, and gain key insights into the current industry.

Now, they’re sharing their top takeaways regarding the advancements in artificial intelligence (AI)! 

The buzzword of the year has seemingly permeated every aspect of our lives and this year’s AFSA conference was no exception. So the question that was posed at AFSA this year is “How does AI fit in the financial world?” We’re here to tell you the good, the bad, and the ugly… 
 

The good:

AI can be used for an endless amount of possibilities. The first thought that people gravitate toward is to use AI as a customer service tool so that customers can get quick, real-time responses to their inquiries. A secondary subject that also gained a lot of interest was to use AI as a data processing tool, in particular, customer tendencies and analyzing the effectiveness of advertising. Lastly, using AI as a research tool to assist businesses in becoming more efficient. As you will read below, these benefits are not without certain risks.
 

The bad:

AI pools its information from the data it has access to. So if the data is inaccurate or undesirable, then the results will be too. If you feed AI junk, it will generate junk. There is concern that AI may not necessarily be a self-driving solution. This also raises several questions: “Should there be disclosures to customers when they are not interacting with a human?” and “Should there be human oversight during a customer’s interaction with AI?” As of now, that seems to be the case. It is too early to just throw artificial intelligence into the world without some vetting and oversight.

The Consumer Financial Protection Bureau (CFPB) met with attendees, and while the topic was more about installment lending (short-term, 6-week programs), know that the CFPB is tracking the use of AI when used in marketing, and it could result in unlawful discrimination. The CFPB believes that AI tools can turbocharge fraud and automate discrimination.  
 

The ugly:

Unconscious bias. As previously mentioned, AI pools its data from what is made accessible to it. When the pool of information it draws from is open-ended and includes biases, stereotypes, and false information, how do you prevent AI from 1) gravitating towards those sources, 2) failing to differentiate fact from fiction, and 3) developing its own biases in the same as a human would?

---

Our team is constantly monitoring this topic. If you have any questions about this topic or Weltman’s consumer collections and/or bankruptcy recovery solutions, connect with Sara and/or Milos at any time. 

This blog is not a solicitation for business, and it is not intended to constitute legal advice on specific matters, create an attorney-client relationship or be legally binding in any way.