Abstract
Maximizing savings, e.g., by choosing an appropriate form of payment, voucher/offer, merchant, etc. is difficult due to the need to analyze a large amount of information to determine the best combination when making a purchase. This disclosure describes a digital payment concierge that can answer complex financial queries. The concierge is implemented using an agent-based framework. To understand user intent, an orchestration agent is implemented using a suitable artificial intelligence model and a custom prompt. The orchestration agent routes tasks to specialized subagents, such as a savings maximizer which finds the best price by stacking offers and considering the user's context, and a fraud detector to identify potential scams. The agents can use tools to gather information such as user profile (with user permission), merchant offers, payment card details, etc. The savings maximizer calculates the effective price by stacking the available offers and provides ranked options and purchase guidance to the user. This framework simplifies maximizing savings and detecting fraud.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Aggarwal, Ankit; Estevez, Jose; Rungsangthiwakorn, Chin Powit; and Dhillon, Samreen, "AI-based Digital Concierge for Complex Financial Queries", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/9050