Abstract: Disclosed are various embodiments for a single tap payment transaction. First, a user of a computing device can be authenticated. Then, in response to authentication of the user, the user can be prompted to select a payment credential. Next, the payment credential and a digital identity document associated with the user can be provided to a payment terminal in data communication with the computing device through an NFC radio. Subsequently, a receipt for a transaction can be received from the payment terminal, the receipt indicating that payment for the transaction was made with the payment credential and that an identity of the user was verified with the digital identity document.

Abstract: Disclosed are various embodiments for automatic enrollment in a merchant rewards program. A point-of-sale (POS) terminal can establish a near field communication (NFC) session with a client device to communicate with the NFC enabled client device. The PoS terminal can then receive a payment instruction for a transaction from the client device, the payment instruction comprising a unique user identifier for a user of the client device. Next, the PoS terminal can determine, based at least in part on the unique user identifier, that a user of the client device is not enrolled in a merchant rewards program. Subsequently, the PoS terminal can send to the client device a request to enroll in the merchant rewards program. Then, the PoS terminal can receive, from the client device, an enrollment authorization. Finally, the PoS terminal can enroll a user of the client device in the merchant rewards program.

Abstract: Disclosed are various embodiments for an NFC-based self-checkout system. Embodiments can include a computing device comprising a near-field communication (NFC) reader. The computing device can obtain, via the NFC reader, a stock keeping unit (SKU) identifier for an item from an item NFC tag attached to the item. It can also obtain, via the NFC reader, a shopping cart identifier and a terminal identifier from a cart NFC tag attached to the shopping cart. It can then receive a user input indicating that a purchase of the item is to be completed. Subsequently, the computing device can establish a secure connection with a payment terminal service associated with the terminal identifier and send the SKU for the item, a payment instrument, and the shopping cart identifier to the payment terminal service.

Abstract: Disclosed are various embodiments for securely and efficiently transmitting various types of digital content (e.g., images, audio, video, etc.) with added assurance regarding the integrity and similarity to the original form of the content. Content to be transmitted can be encoded into a latent encoded representation of the content using a trained AI encoder, thereby reducing the size or compressing the content. A digital signature can also be applied to the encoded representation. The sender can send the original content and the signed encoded representation to a receiver. The receiver can encode the received content using the AI encoder and compare the encoded representations to determine similarities. In addition, the receiver side can validate the digital signature of the received encoded representation. The comparison of the encoded representations and the validation of the digital signature can both be used to validate the integrity of the received content.

Abstract: Disclosed are various embodiments for an encrypted context-based prompt system for interactions between multiple entities. In one embodiment, a system comprising a computing device configured to receive a request to activate a prompt session. The request comprises a template identifier representing a type of interaction between a first device and a second device. A decrypted term is determined by decrypting an encrypted term associated with the request. A large language model prompt is generated based at least in part on the template identifier and the decrypted term. A proposed term is generated by executing a large language model application based at least in part on the large language model prompt. An encrypted proposed term is transmitted to the first client device or the second client device.

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.

Abstract: Disclosed are various embodiments for implementing a decentralized messaging system that allows users to exchange messages via an inbox maintained through a distributed ledger. A messaging inbox can be created for a user in a distributed ledger using an inbox address that includes a string of alphanumeric numbers without an identifying domain name or identifying information about the user. Furthermore, a user may define content preferences associated with the inbox that can be used by the distributed ledger to filter content received in the inbox to minimize unwanted or undesired content provided to the user.

Abstract: Disclosed are various embodiments for long term context-based small language model aided authentication. Various embodiments of the present disclosure can receive a contextual instruction associated with a credential. The contextual instruction can represent a category of input that is accepted to obtain a user credential. Various embodiments can train a small language model to generate an authentication small language model such that the authentication small language model returns the credential in response to determining that an input is within the category of input of the contextual instruction. Various embodiments can then receive a client input and prompt the authentication small language model to make a determination regarding the client input is within the category of input of the contextual instruction.

Abstract: Disclosed are various embodiments for dynamic enforcement of large language model prompt templates and prompt template categorization. In one example, a system comprise a computing device that is configured to identify a prompt that has been submitted by a client device for a large language model (LLM) service and determine that the prompt fails to match an existing prompt template. The prompt and an unidentified prompt are determined to meet a similarity threshold based at least in part on a common prompt component shared between the prompt and the unidentified prompt. A prompt template is generated for the LLM service based at least in part on the prompt and the unidentified prompt meeting the similarity threshold, the prompt template comprising the common prompt component and a placeholder.

Abstract: Disclosed are various approaches for language model based options selections. In one example, a service can generate a user interface that identifies user interactions received from a client device to register transaction instruments. Transaction facilitation websites can be scraped to identify transaction instrument terms for the registered transaction instruments. An option selection language model can be generated and trained to select a subset of the registered transaction instruments. The option selection language model, the transaction instrument data, and the transaction instrument terms can be transmitted to the client device.

Abstract: Disclosed are various embodiments for protecting payment credentials. A request for a double-encrypted token payload (dETP) can be received from a payment initiator, the request comprising a merchant identifier and an amount of a transaction. An encrypted token payload (ETP) that represents the payment request and a payment credential can be prepared, the ETP being encrypted with a first encryption key stored on the computing device. The ETP can then be sent to a client application executing on a client device associated with an owner of the payment credential. The dETP can be received from the client application executing on the client device, the dETP being encrypted with a second encryption key stored on the client device. The dETP can then be returned to the payment initiator.

Abstract: Disclosed herein are system, method, and computer program product embodiments for autonomously repairing software by leveraging a large language model (LLM). A control system may detect a first error associated with an application executing in a region. The control system may then repair the first error associated with the application by: identifying a source of the first error within the application; generating a solution by inputting the source of the first error to an LLM; and implementing the solution via the LLM. The control system may then determine that the application is repaired by: executing the application; generating an output; and comparing the output to a predefined value. The control system may then deploy the application in the region.

Abstract: Disclosed are various embodiments for an encrypted context-based prompt system for interactions between multiple entities. In one embodiment, a system comprising a computing device configured to receive a request to activate a prompt session. The request comprises a template identifier representing a type of interaction between a first device and a second device. A decrypted term is determined by decrypting an encrypted term associated with the request. A large language model prompt is generated based at least in part on the template identifier and the decrypted term. A proposed term is generated by executing a large language model application based at least in part on the large language model prompt. An encrypted proposed term is transmitted to the first client device or the second client device.

Abstract: Disclosed are various embodiments for the tokenized combination of accounts prior to a transaction. A first account identifier is obtained for a first transaction card. Then, a split transaction account creation request is sent to a split transaction service, the split transaction account creation request comprising the first account identifier representing the first transaction card and a second account identifier representing a second transaction card. A split transaction token is received from the split transaction service, the split transaction token representing a virtualized transaction card for use in a transaction. Subsequently, the split transaction token is provided to a point-of-sale (PoS) device to complete a transaction.

Abstract: Disclosed are various embodiments for a return management service that acts as an intermediary between a user (e.g., a consumer) and a merchant to intelligently recommend and initiate returns and/or refunds for the user. According to various examples, the return management service includes a large language model (LLM) that can learn from information associated with transactions, returns, and/or refunds and obtained from users, merchants, and/or issuers to proactively update, personalize for a user, provide return/refund recommendations, and carry out communications with merchants and/or issuers in line with a user's request. Users can register to participate with the return management service so that when a user wishes to return an item and/or request a refund for a purchased item or service, the user can receive a recommendation from the return management service for returning the items and/or receiving a refund.

Abstract: Examples herein can receive a request to authorize a card payment from a point-of-sale device. The request is associated with an attempted card payment. The system further determines a user account associated with the payment request and identifies at least one passive authentication factor linked to the user account. The passive authentication factor is associated with at least one device associated with the user account. The system then determines whether the passive authentication factor is satisfied by analyzing the device. If the passive authentication factor is satisfied, the system transmits a transaction authorization indication to the point-of-sale device.

Abstract: Disclosed are various embodiments for generating payment accessibility notifications for individuals with impairments. In one non-limiting example, a system includes a first computing device configured to identify a second computing device using a wireless communication protocol in a store location. A first notification for the first computing device is activated based at least in part on an accessibility type. A user interface is displayed for receiving client data, and the client data includes a product entered by a user. An expected amount is determined for the product based at least in part on the client data and transmitted to the second computing device. A pending amount for the product is received from the second computing device. A second notification is activated for the first computing device prior to a completion of the purchase based at least in part on the pending amount and the expected amount.

Abstract: Disclosed are various embodiments for using decentralized identity to verify account information for a user requesting a transfer of funds between one or more financial entities (e.g., banks, credit unions, brokerage firms, mortgage companies, etc.). A user having account with a first entity can request that a balance credential be created claiming that the balance of funds within the account meets or exceeds a predefined threshold. The balance credential can be created and saved on a distributed ledger in relation to a decentralized identity identifier (DID) of the user. When a second entity wants to verify the account balance of the user, the verifying entity which can use the DID to access the balance credential on the distributed ledger and validate the accessed credential.

Abstract: Disclosed are various embodiments for generating payment accessibility notifications for individuals with impairments. In one non-limiting example, a system includes a first computing device configured to identify a second computing device using a wireless communication protocol in a store location. A first notification for the first computing device is activated based at least in part on an accessibility type. A user interface is displayed for receiving client data, and the client data includes a product entered by a user. An expected amount is determined for the product based at least in part on the client data and transmitted to the second computing device. A pending amount for the product is received from the second computing device. A second notification is activated for the first computing device prior to a completion of the purchase based at least in part on the pending amount and the expected amount.

Abstract: Disclosed are various embodiments for a service for decentralized browser based wallets. Various embodiments of the present disclosure can receive a first passcode and a second passcode from a security provider device. Various embodiments can use a cryptowallet to decrypt a first encrypted private key using the first passcode to generate a decrypted private key. Various embodiments can use a cryptowallet to sign a transaction request with the decrypted private key. Various embodiments can use a cryptowallet to generate a second encrypted private key by encrypting the decrypted private key using the second passcode.