Encrypted autonomous agent execution using cross-verification methods in distributed systems

Inventors: Ganesh Prasad Bhat (New Jersey, NJ); James Myers (Clearwater, FL)

Assignee: CITIBANK, N.A.

H04L47/822 H04L67/1097

View Patent ↗

Loading inventors, assignments & file history…

Monitor This Case

Get email alerts when status or documents change.

Order Certified Copies

Most orders are placed with the USPTO same day — all within 24 business hours.

Order via The Patent Place →

Pre-filled with this patent's details

Quick Facts

Patent No.: US 12706859
App. No.: 19/334,759
Granted: Aug 11, 2026
Kind: B1

External Links

USPTO Patent Center↗ Full Patent PDF↗

Abstract

Systems and methods are disclosed comprising techniques for agentic cross-verification in distributed systems, such as obtaining a request to validate a decision state for a first memory structure of the first multi-agent storage that is accessible to a first AI agent set, transmitting a verification artifact representing the decision state of the first multi-agent storage to a second multi-agent storage that stores the verification artifact, receiving a verification status of the verification artifact that is determined by the second AI agent set, generating a first programmatic workflow set configured to trigger a first sequence of computer-executable commands at the first memory structure, generating a second programmatic workflow set configured to trigger a second sequence of computer-executable commands at the second memory structure, and automatically executing the first programmatic workflow set using the first AI agent set and the second programmatic workflow set using the second AI agent set.

Claims (87)

1 . A system for performing privacy-preserving cross-verification of memory storage states, the system comprising:

at least one hardware processor; and

at least one non-transitory memory storing instructions, which, when executed by the at least one hardware processor, cause the system to:

obtain, via a first multi-agent storage, a request to validate a decision state for a first memory structure of the first multi-agent storage that is accessible to a first AI agent set,

wherein the decision state indicates a recorded artifact generated from executing a prior programmatic workflow set via the first AI agent set;

transmit a verification artifact representing the decision state of the first multi-agent storage to a second multi-agent storage that is configured to store the verification artifact,

wherein the verification artifact is generated by applying a transformation operation set on the recorded artifact of the decision state, and

wherein the second multi-agent storage comprises a second memory structure accessible to a second AI agent set;

receive, via the second multi-agent storage, a verification status of the verification artifact that is determined by the second AI agent set; and

responsive to the verification status indicating validation of the decision state for the first multi-agent storage:

generate, using the first AI agent set, a first programmatic workflow set configured to trigger a first sequence of computer-executable commands at the first memory structure of the first multi-agent storage;

generate, using the second AI agent set, a second programmatic workflow set configured to trigger a second sequence of computer-executable commands at the second memory structure of the second multi-agent storage; and

automatically cause execution of the first programmatic workflow set via the first AI agent set and the second programmatic workflow set via the second AI agent set.

2 . The system of claim 1 further caused to:

assign, prior to execution the first programmatic workflow set and the second programmatic workflow set, a first restriction status on the first memory structure of the first multi-agent storage and a second restriction status on the second memory structure of the second multi-agent storage,

wherein the first restriction status prevents modification to the first memory structure until completion of the first sequence of computer-executable commands,

wherein the second restriction status prevents modification to the second memory structure until completion of the second sequence of computer-executable commands.

3 . The system of claim 1 further caused to:

automatically generate a verification record including a representation of the verification status for the verification artifact; and

transmit the verification record to the first multi-agent storage and the second multi-agent storage,

wherein the verification record is stored in associated with respective identifier strings identifying the first AI agent set and the second AI agent set.

4 . The system of claim 1 , wherein the verification status indicates invalidation of the decision state for the first multi-agent storage, and wherein the system is further caused to:

transmit a notification message to a predetermined network address indicating the verification status for the verification artifact.

5 . The system of claim 1 ,

wherein the verification artifact includes a zero-knowledge proof generated by the first AI agent set,

wherein the zero-knowledge proof indicates that the decision state for the first memory structure of the first multi-agent storage satisfies an internal operational dataset of the second AI agent set, and

wherein the verification status is determined using the zero-knowledge proof.

6 . The system of claim 1 ,

wherein the first AI agent set comprises a first AI agent subset and a second AI agent subset,

wherein the first AI agent subset comprises AI agents that generate the recorded artifact via executing the prior programmatic workflow set, and

wherein the second AI agent subset comprises AI agents that generate the first programmatic workflow set.

7 . The system of claim 1 , wherein the first AI agent set and the second AI agent set are configured to execute the first programmatic workflow set and the second programmatic workflow set contemporaneously.

8 . The system of claim 1 ,

wherein the first sequence of computer-executable commands of the first programmatic workflow set comprises at least one first command for transferring machine-readable data from the first multi-agent storage to the second multi-agent storage, and

wherein the second sequence of computer-executable commands comprises at least one second command to process the machine-readable data received from the first multi-agent storage.

9 . The system of claim 1 , wherein the transformation operation set comprises at least one of: a cryptographic hash function, a keyed hash function, or a deterministic encoding function.

10 . The system of claim 1 , wherein the first multi-agent storage is a first distributed ledger and the second multi-agent storage is a second distributed ledger.

11 . A non-transitory, computer-readable storage medium comprising instructions recorded thereon, wherein the instructions, when executed by at least one data processor of a system, cause the system to: