US-20260127631-A1 - GENERATIVE ARTIFICIAL INTELLIGENCE UPSELL ENGINE

US20260127631A1US 20260127631 A1US20260127631 A1US 20260127631A1US-20260127631-A1

Abstract

Disclosed herein are system, method, and computer program product embodiments for using generative AI to support an upsell engine. A point-of-sale (POS) system may combine transaction data, merchant data, and consumer data into a shared embedding space comprising a numerical matrix. A machine learning model in connection with the POS system may generate an offer based on the shared embedding of the transaction data, consumer data, and merchant inventory data. The offer may be unique to the transaction. The POS system may prevent merchant data from being transmitted to a banking server, and may prevent consumer data from being transmitted to a merchant server. The POS system may output the offer.

Inventors

Alaric M. Eby
ANDRAS L. FERENCZI

Assignees

AMERICAN EXPRESS TRAVEL RELATED SERVICES COMPANY, INC.

Dates

Publication Date: 20260507
Application Date: 20241104

Claims (20)

1 . A computer implemented method, the method comprising: combining, at a point-of-sale (POS) system, transaction data, merchant data, and consumer data into a shared embedding space comprising a numerical matrix; generating, by a machine learning model in connection with the POS system, an offer based on the shared embedding space of the transaction data, the consumer data, and the merchant data, wherein the offer is unique to the transaction; preventing, by the POS system, the merchant data from being transmitted to a banking server; preventing, by the POS system, the consumer data from being transmitted to a merchant server; and outputting the offer via the POS system.
2 . The computer implemented method of claim 1 , further comprising generating a preliminary offer at the POS system when an item is scanned, wherein the preliminary offer is generated based on the scanned item and the merchant data.
3 . The computer implemented method of claim 1 , wherein prior to outputting the offer, validating the offer by: determining a discount in the offer is less than a maximum allowable discount parameter; determining an item in the offer is eligible for the offer; and determining a consumer identified in the offer is eligible, wherein the determination is made using the consumer data.
4 . The computer implemented method of claim 1 , wherein the offer is further based on a preliminary offer, wherein generating the preliminary offer comprises; receiving, at the POS system, first sensor data; generating the preliminary offer by the machine learning model in connection with the POS system, using the first sensor data and the merchant data; receiving, at the POS system, second sensor data; and updating the preliminary offer by the machine learning model in connection with the POS system, using the first sensor data, the second sensor data, and the merchant data.
5 . The computer implemented method of claim 4 , wherein prior to receiving the first sensor data, the method further comprises: detecting, by the POS system, a client device of a consumer by at least one of a software application on the client device or a sensor; receiving, at the POS system, approval from a client device to track the consumer with a sensor; and in response to receiving approval from the client device, transmitting, by the POS system, a message to the sensor to track the consumer of the client device.
6 . The computer implemented method of claim 4 , wherein the first sensor data is generated by a first camera, wherein the first sensor data includes first video data indicating a first item the consumer interacted with in a store, wherein the second sensor data is generated by a second camera, and wherein the second sensor data includes second video data indicating a second item the consumer interacted with in the store.
7 . The computer implemented method of claim 4 , wherein the first sensor data and second sensor data include data from at least one of a camera, microphone, motion sensor, pressure sensors, GPS, RFID sensor, or proximity sensor.
8 . A system, comprising: a memory; and at least one processor coupled to the memory and configured to: combine transaction data, merchant data, and consumer data into a shared embedding space comprising a numerical matrix; generate, by a machine learning model, an offer based on the shared embedding space of the transaction data, the consumer data, and the merchant data, wherein the offer is unique to the transaction; prevent the merchant data from being transmitted to a banking server; prevent the consumer data from being transmitted to a merchant server; and output the offer.
9 . The system of claim 8 , wherein the at least one processor is further configured to generate a preliminary offer at the POS system when an item is scanned, wherein the preliminary offer is generated based on the scanned item and the merchant data.
10 . The system of claim 8 , wherein prior to outputting the offer, the at least one processor is further configured to validate the offer by: determine a discount in the offer is less than a maximum allowable discount parameter; determine an item in the offer is eligible for the offer; and determine a consumer identified in the offer is eligible, wherein the determination is made using the consumer data.
11 . The system of claim 8 , wherein the offer is further based on a preliminary offer, wherein to generate the preliminary offer the at least one processor is further configured to: receive first sensor data; generate the preliminary offer by the machine learning model using the first sensor data and the merchant data; receive second sensor data from the sensor; and update the preliminary offer by the machine learning model using the first sensor data, the second sensor data, and merchant data.
12 . The system of claim 11 wherein prior to receiving the first sensor data, the at least one processor is further configured to: detect a client device of a consumer by at least one of a software application on the client device or a sensor; receive approval from a client device to track the consumer with a sensor; and in response to receiving approval from the client device, transmit a message to the sensor to track the consumer of the client device.
13 . The system of claim 11 , wherein the first sensor data is generated by a first camera, wherein the first sensor data includes first video data indicating a first item the consumer interacted with in a store, wherein the second sensor data is generated by a second camera, and wherein the second sensor data includes second video data indicating a second item the consumer interacted with in the store.
14 . The system of claim 11 wherein the first sensor data and second sensor data include data from at least one of a camera, microphone, motion sensor, pressure sensors, GPS, RFID sensor, or proximity sensor.
15 . A non-transitory computer-readable device having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising: combining, at a point-of-sale (POS) system, transaction data, merchant data, and consumer data into a shared embedding space comprising a numerical matrix; generating, by a machine learning model in connection with the POS system, an offer based on the shared embedding space of the transaction data, the consumer data, and the merchant data, wherein the offer is unique to the transaction; preventing, by the POS system, the merchant data from being transmitted to a banking server; preventing, by the POS system, the consumer data from being transmitted to a merchant server; and outputting the offer via the POS system.
16 . The non-transitory computer-readable device of claim 15 , wherein the operations further comprise generating a preliminary offer at the POS system when an item is scanned, wherein the preliminary offer is generated based on the scanned item and the merchant data.
17 . The non-transitory computer-readable device of claim 16 , wherein the offer is further based on a preliminary offer and wherein the operations further comprise: receiving, at the POS system, first sensor data; generating the preliminary offer by the machine learning model in connection with the POS system, using the first sensor data and the merchant data; receiving, at the POS system, second sensor data; and updating the preliminary offer by the machine learning model in connection with the POS system, using the first sensor data, the second sensor data, and the merchant data.
18 . The non-transitory computer-readable device of claim 17 , wherein prior to receiving the first sensor data, the operations further comprise: detecting, by the POS system, a client device of a consumer by at least one of a software application on the client device or a sensor; receiving, at the POS system, approval from a client device to track the consumer with a sensor; and in response to receiving approval from the client device, transmitting, by the POS system, a message to the sensor to track the consumer of the client device.
19 . The non-transitory computer-readable device of claim 17 , wherein the first sensor data is generated by a first camera, wherein the first sensor data includes first video data indicating a first item the consumer interacted with in a store, wherein the second sensor data is generated by a second camera, and wherein the second sensor data includes second video data indicating a second item the consumer interacted with in the store.
20 . The non-transitory computer-readable device of claim 17 , wherein the first sensor data and second sensor data include data from at least one of a camera, microphone, motion sensor, pressure sensors, GPS, RFID sensor, or proximity sensor.

Description

BACKGROUND Field This field is generally related to increasing data security of systems interacting with a point of sale (POS) system using generative artificial intelligence (AI) to enable an upsell engine that maintains the privacy of information between the systems. Related Art Concerns surrounding data security and privacy are increasing given: (1) the increase in amount of data generated and stored; (2) the number of systems utilizing that data; and (3) the interconnectedness of those systems. For example, a merchant may inadvertently share a consumer’s credit card or other banking information with a third party. Alternatively, a bank may experience a cyberattack, causing merchant and customer data to be leaked to nefarious third parties. In addition to security, there are also proprietary concerns regarding sharing data. An entity may spend significant resources and capital building a data set that accurately captures its products, services, customers, etc. For example, a financial institution that collects consumer data may not wish to share this data based on: (1) regulatory restrictions; and (2) the cost of creating the data set. However, the combination of data from different entities may be used to generate insights regarding likely successful products and expected consumer behavior. Thus, there is a need to increase the ability to securely share data, in real-time, between multiple parties. BRIEF SUMMARY Disclosed herein are system, apparatus, device, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for using generative AI to support an upsell engine with limited transmission of customer and merchant data between components within a point of sale (POS) system. This disclosure describes a POS system that leverages a trained machine learning model to generate unique offers based on information provided by merchant systems (e.g., merchant proprietary information, shopping cart information), consumer devices (e.g., consumer information and/or behavior) and bank backend systems (e.g., bank transaction history). The POS system may leverage both banking information from bank backend systems and merchant information, while maintaining privacy and separation of bank and merchant information by limiting communications between the bank backend system and merchant system. The POS system may combine the relevant bank, merchant, and transaction data, input it to a trained machine learning model, and generate a custom offer. The custom offer may be provided to the consumer involved in the purchase. The POS system described herein improves network and data privacy by allowing merchant and bank information to be leveraged, without having to share it with untrusted parties. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings are incorporated herein and form a part of the specification. FIG. 1 depicts a block diagram of a merchant environment, according to some embodiments. FIG. 2 depicts a block diagram illustrating a method for utilizing AI for real-time unique offer generation, according to some embodiments. FIG. 3 depicts a flowchart illustrating a method for utilizing AI for real-time unique offer generation, according to some embodiments. FIG. 4 depicts a flowchart illustrating a method for leveraging sensor data for offer generation, according to some embodiments. FIG. 5 depicts an example computer system useful for implementing various embodiments. In the drawings, like reference numbers generally indicate identical or similar elements. Additionally, generally, the left-most digit(s) of a reference number identifies the drawing in which the reference number first appears. DETAILED DESCRIPTION Provided herein are system, apparatus, device, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for using generative artificial intelligence (AI) to enable an upsell engine. A machine learning model may be leveraged to predict real-time offers based on combined bank and merchant information while maintaining the privacy of bank and merchant information by limiting communications between these systems. Merchants and banks may both generate and send offers to customers. For example, a bank that offers a credit card may provide offers to credit card users based off of various factors such as credit limit, credit utilization, and purchase history. Merchants may also generate and transmit offers to customers based off of loyalty programs, purchase history, and ongoing promotions. However, banking systems and merchant servers often distribute these offers without having access to data from the other party. For example, a bank may offer a 10% discount at a merchant to all credit card holders. However, an offer of this type fails to take into account the merchant’s data. As discussed above, a bank may not wish to share any consumer data with a merchant, resulting in a merchant being unable to understand their cust