Under 1% of Credit Unions Could Run an Agent on Their Data

Ankur Patel interviews Garrett Jorewicz, who leads the Innovation and Enterprise Solutions team at Credit Union 1 (CU1).<br /><br />📩 Subscribe to the Main Street AI Newsletter: https://multimodal.beehiiv.com/subscribe?utm_source=podcast&amp;utm_medium=description&amp;utm_campaign=podcast<br /><br />Credit Union 1 grew from $1.3 billion to $2.6 billion in assets through eight to ten mergers a year and doubled its members without adding a single person to its contact center. Garrett's argument is that this was the easy part: he puts the share of credit unions that could run an AI agent against all their data at under 1%, CU1 included, and says getting there means freezing projects to fix the data layer first.<br /><br />Guest: Garrett Jorewicz, Innovation and Enterprise Solutions, Credit Union 1 (CU1)<br />Host: Ankur Patel, CEO &amp; Founder, Multimodal<br /><br />0:00 Introduction: Meet Garrett<br />0:21 Under 1% of credit unions<br />0:48 Why the data still isn't in order<br />3:37 Voluntary open heart surgery<br />5:41 Contact center as the entry point<br />6:59 Eight to ten mergers a year<br />13:08 Four AI vendors, four knowledge bases<br />15:31 The contact center RFP<br />17:04 The data cockpit<br />19:04 Sorting through vendor noise<br />23:22 Who pays for AI<br />27:44 Do I need to replace that person<br />29:11 Same vendors, same tools<br />34:05 Friction drives behavior<br />37:03 Twelve seconds or eleven minutes on hold<br />38:55 Credit unions in ten years<br />41:06 What credit unions should stop doing<br />43:42 Closing