IP Library › Granted Patent US 12,542,776
Granted Patent B2
US 12,542,776 · App. 18/225,200 · Granted Feb 3, 2026

Verifying human interaction through an encrypted channel

Inventors: Manu Kurian (Dallas, TX); Ana Maxim (Arlington, VA); Vinesh Patel (London, GB); Michael Young (Davidson, NC)
Assignee: Bank of America Corporation
H04L63/0861H04L63/0428
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 12,542,776
App. No.
18/225,200
Granted
Feb 3, 2026
Kind
B2
Abstract

Apparatus and methods to verify human interaction through an encrypted channel are provided. A human verification program may create an encrypted channel between two or more devices when a communication between users of the devices is started. Data may be transferred over the encrypted channel when a user performs a verifiably human action. The data may be analyzed to determine whether the user is more likely than not a bot. The channel may be collapsed when a pre-determined condition is satisfied.

Claims (62)

1 . A human verification computer program product, the computer program product comprising executable instructions, the executable instructions when executed by a processor on a computer system:

create an encrypted channel between the computer system and a second device when a first user and a second user begin a communication over a network through the computer system and the second device;

receive data from the second device over the encrypted channel;

analyze the data to verify that the second user is a human; and

collapse the encrypted channel when a pre-determined condition is satisfied;

wherein:

the data is created by the user of the second device performing at least one verifiably human action; and

wherein the verifiably human action comprises answering an arithmetic question incorrectly and placing a particular weight on a weight sensor electronically coupled to a transmitter;

the first user uses the computer system; and

the second user uses the second device.

2 . The human verification computer program product of claim 1 wherein the verifiably human action further comprises responding to a captcha.

3 . The human verification computer program product of claim 1 wherein the verifiably human action further comprises measuring a parameter with a biometric sensor.

4 . The human verification computer program product of claim 3 wherein the parameter is body temperature.

5 . The human verification computer program product of claim 3 wherein the parameter is pulse rate.

6 . The human verification computer program product of claim 1 wherein the verifiably human action further comprises physically connecting two or more components to form an electrical circuit.

7 . The human verification computer program product of claim 1 wherein the analysis applies one or more filtering rules to determine whether the user is a human.

8 . The human verification computer program product of claim 1 wherein the verifiably human action further comprises typing a supplied random phrase.

9 . The human verification computer program product of claim 8 wherein the analysis analyzes one or more of:

a) typing cadence;

b) typing speed;

c) typing accuracy;

d) key pressure; and

e) a combination of a)-d).

10 . The human verification computer program product of claim 1 wherein the pre-determined condition is either the first user or the second user ending the communication.

11 . The human verification computer program product of claim 1 wherein the pre-determined condition is an elapsed time.

12 . The human verification computer program product of claim 1 wherein the pre-determined condition is when the analysis concludes that both users are more likely than not human.

13 . The human verification computer program product of claim 1 wherein the pre-determined condition is when the analysis concludes that either the first user or the second user is more likely than not a bot.

14 . The human verification computer program product of claim 13 wherein the instructions further notify one user that the other user is more likely than not a bot.

15 . The human verification computer program product of claim 1 wherein the network is an internal network.

16 . The human verification computer program product of claim 1 wherein the analysis utilizes one or more artificial intelligence/machine learning (“AI/ML”) algorithms.

17 . An apparatus for human verification, the apparatus comprising:

two or more network nodes, each network node comprising:

a node communication link;

a node processor; and

a node non-transitory memory configured to store at least:

a node operating system; and

a human verification application;

wherein the human verification application on a first network node:

creates an encrypted channel between the first network node and a second network node when a user of the first network node and a user of the second network node begin a communication over a network through the first and second network nodes;

receives data from the second network node over the encrypted channel, wherein the data is created by the user of the second network node performing a verifiably human action; and

wherein the verifiably human action comprises physically connecting two or more components to form an electrical circuit and placing a particular weight on a weight sensor electronically coupled to a transmitter;

analyzes the data to verify that the second user is a human; and

collapses the encrypted channel when a pre-determined condition is satisfied.

18 . A method for human verification, the method comprising the steps of:

creating, by a human verification program on a central server, an encrypted channel between:

the central server;

a first network node; and

a second network node

when a first user of the first network node and a second user of the second network node begin a communication over a network through the first and second network nodes;

receiving, at the human verification program, first data from the first network node over the encrypted channel, wherein the data is created by the first user performing a verifiably human action; and

wherein the verifiably human action comprises typing a supplied random phrase and placing a particular weight on a weight sensor electronically coupled to a transmitter;

receiving, at the human verification program, second data from the second network node over the encrypted channel, wherein the data is created by the second user performing a verifiably human action;

analyzing the first data to verify that the first user is a human, wherein the analysis analyzes one or more of:

a) typing cadence;

b) typing speed;

c) typing accuracy;

d) key pressure; and

e) a combination of a)-d);

analyzing the second data to verify that the second user is a human;

informing the first user when the second user is more likely than not a bot;

informing the second user when the first user is more likely than not a bot; and

collapsing the encrypted channel when a pre-determined condition is satisfied.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 24, 2023
From: KURIAN, MANU; MAXIM, ANA; PATEL, VINESH; YOUNG, MICHAEL
To: BANK OF AMERICA CORPORATION
Reel/Frame 064352/0155 →
Continuity (1)
Related Publication 20250039170A1 · Jan 30, 2025
References Cited (7)
US 8782425B2 · Steeves · 2014 [cited by examiner]
US 11675872B2 · Holloway · 2023 [cited by examiner]
US 11695771B2 · Natarajan · 2023 [cited by examiner]
US 12101317B2 · Duthoit · 2024 [cited by examiner]
US 12143370B1 · Ou · 2024 [cited by examiner]
US 20120197688A1 · Townshend · 2012 [cited by examiner]
US 20170093864A1 · Pestun · 2017 [cited by examiner]