# US 2026/0100959 A1 · Security action based on communication-based analysis

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0100959 A1
- Document ID: US20260100959A1
- Status: Published application
- Kind: A1
- Application number: 18/908,220
- Filed: 2024-10-07
- Published: 2026-04-09
- Earliest claimed: 2024-10-07
- Assignees: Microsoft Technology Licensing, LLC
- Applicants: Microsoft Technology Licensing, LLC
- Inventors: Doron BAR SHALOM; Andrey KARPOVSKY; Fady COPTY
- Web page: https://patentgrep.com/patent/US20260100959A1

## CPC

- G06F 21/554: {involving event detection and direct action}
- H04L 63/1416: {Event detection, e.g. attack signature detection}
- H04L 41/16: using machine learning or artificial intelligence
- H04L 63/1433: {Vulnerability analysis}
- G06F 21/552: {involving long-term monitoring or reporting}
- H04L 63/1425: {Traffic logging, e.g. anomaly detection}
- H04L 63/1441: {Countermeasures against malicious traffic (countermeasures against attacks on cryptographic mechanisms H04L9/002)}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Techniques are described herein that are capable of performing a security action based on a communication-based analysis. A security event, which is triggered by an operation performed by a user in an organization, is detected. A security analysis result is generated by determining whether a communication history of the user includes (1) a communication from the user that initiates an interaction with another user in the organization and/or a communication that is addressed specifically to the user from another user in the organization, (2) a communication from the user that references the operation, and/or (3) a communication that provides an explanation of a purpose of the operation that satisfies an explanation criterion. In response to the security event, a security action is performed with regard to the operation as a result of the security analysis result satisfying a security criterion.

## Claims

### Claim 1

A system comprising:
a processor system; and
a memory that stores computer-executable instructions that are executable by the processor system to at least:
detect a security event triggered by an operation that is performed by a user in an organization;
generate a first security analysis result by determining whether a communication history of the user comprises at least one of a first communication from the user that initiates an interaction with another user in the organization or a second communication that is addressed specifically to the user from another user in the organization;
generate a second security analysis result by determining whether the communication history of the user comprises a third communication from the user that references the operation;
generate a third security analysis result, which comprises determining whether the communication history of the user comprises a fourth communication that provides an explanation of a purpose of the operation that satisfies an explanation criterion, using an artificial intelligence (AI) model; and
in response to the security event, trigger execution of an instruction, which causes a security action to be performed with regard to the operation as a result of a combination of the first, second, and third security analysis results satisfying a security criterion.

### Claim 2 (depends on 1)

The system of claim 1, wherein the computer-executable instructions are executable by the processor system to at least:
generate the first security analysis result by determining at least one of a number of first communications in the communication history that are from the user and that initiate an interaction with another user in the organization or a number of second communications in the communication history that are addressed specifically to the user from another user in the organization.

### Claim 3 (depends on 1)

The system of claim 1, wherein the computer-executable instructions are executable by the processor system to at least:
generate the second security analysis result by determining a number of third communications in the communication history that are from the user and that reference the operation.

### Claim 4 (depends on 1)

The system of claim 1, wherein the computer-executable instructions are executable by the processor system to at least:
generate the third security analysis result by determining whether a value of an attribute of a user-generated communication, which is generated by the user and which references the operation, in the communication history of the user corresponds to a reference value of the attribute of a reference user-generated communication, which references a historical operation that corresponds to the operation performed by the user.

### Claim 5 (depends on 1)

The system of claim 1, wherein the computer-executable instructions are executable by the processor system to at least:
generate the third security analysis result by determining whether a value of an attribute of a system-generated communication, which is generated by a system utilized by the user and which references the operation, corresponds to a reference value of the attribute of a reference system-generated communication, which references a historical operation that corresponds to the operation performed by the user.

### Claim 6 (depends on 1)

The system of claim 1, wherein the computer-executable instructions are executable by the processor system to at least:
generate the third security analysis result by determining whether the fourth communication provides the explanation that satisfies the explanation criterion by determining whether the fourth communication is a request, which is received by the user, to perform the operation and further by determining whether the request corresponds to a business role of the user.

### Claim 7 (depends on 1)

The system of claim 1, wherein the computer-executable instructions are executable by the processor system to at least:
generate embeddings that represent communications in the communication history of the user; and
generate the third security analysis result by causing the AI model to compare the embeddings, which represent the communications in the communication history of the user, to a reference embedding, which represents the explanation criterion; and
wherein an extent to which an embedding of the fourth communication corresponds to the reference embedding indicates whether the fourth communication provides the explanation of the purpose of the operation that satisfies the explanation criterion.

### Claim 8 (depends on 1)

The system of claim 1, wherein the first security analysis result comprises a first security score, which is a numerical representation of whether the communication history of the user comprises at least one of the first communication that initiates the interaction with another user in the organization or the second communication that is addressed specifically to the user from another user in the organization;
wherein the second security analysis result comprises a second security score, which is a numerical representation of whether the communication history of the user comprises the third communication from the user that references the operation;
wherein the third security analysis result comprises a third security score, which is a numerical representation of whether the communication history of the user comprises the fourth communication that provides the explanation of the purpose of the operation that satisfies the explanation criterion; and
wherein the computer-executable instructions are executable by the processor system to at least:
generate a combined security score by combining the first, second, and third security scores; and
cause the security action to be performed with regard to the operation as a result of the combined security score satisfying the security criterion.

### Claim 9 (depends on 1)

The system of claim 1, wherein the computer-executable instructions are executable by the processor system to at least:
generate a first security score, which corresponds to at least one of a number of first communications in the communication history that are from the user and that initiate an interaction with another user in the organization or a number of second communications in the communication history that are addressed specifically to the user from another user in the organization;
generate a second security score, which corresponds to a number of third communications in the communication history that are from the user and that reference the operation;
generate a third security score, which corresponds to an extent to which the explanation provided by the fourth communication satisfies the explanation criterion; and
perform the security action with regard to the operation as a result of a combination of the first, second, and third security scores satisfying the security criterion.

### Claim 10

A method implemented by a computing system, the method comprising:
detecting a security event triggered by an operation that is performed by a user in an organization;
generating a first security analysis result by determining whether a communication history of the user comprises at least one of a first communication from the user that initiates an interaction with another user in the organization or a second communication that is addressed specifically to the user from another user in the organization;
generating a second security analysis result by determining whether the communication history of the user comprises a third communication from the user that references the operation;
causing an artificial intelligence (AI) model to generate a third security analysis result by determining whether the communication history of the user comprises a fourth communication that provides an explanation of a purpose of the operation that satisfies an explanation criterion; and
in response to the security event, triggering execution of an instruction, which causes a security action to be performed with regard to the operation by selecting the security action from a plurality of security actions as a result of a combination of the first, second, and third security analysis results satisfying a security criterion.

### Claim 11 (depends on 10)

The method of claim 10, wherein generating the first security analysis result comprises:
determining at least one of a number of first communications in the communication history that are from the user and that initiate an interaction with another user in the organization or a number of second communications in the communication history that are addressed specifically to the user from another user in the organization.

### Claim 12 (depends on 10)

The method of claim 10, wherein generating the second security analysis result comprises:
determining a number of third communications in the communication history that are from the user and that reference the operation.

### Claim 13 (depends on 10)

The method of claim 10, wherein causing the AI model to generate the third security analysis result comprises:
causing the AI model to determine whether a value of an attribute of a user-generated communication, which is generated by the user and which references the operation, in the communication history of the user corresponds to a reference value of the attribute of a reference user-generated communication, which references a historical operation that corresponds to the operation performed by the user.

### Claim 14 (depends on 10)

The method of claim 10, wherein causing the AI model to generate the third security analysis result comprises:
causing the AI model to determine whether a value of an attribute of a system-generated communication, which is generated by a system utilized by the user and which references the operation, corresponds to a reference value of the attribute of a reference system-generated communication, which references a historical operation that corresponds to the operation performed by the user.

### Claim 15 (depends on 10)

The method of claim 10, wherein causing the AI model to generate the third security analysis result comprises:
causing the AI model to determine whether the fourth communication provides the explanation that satisfies the explanation criterion by determining whether the fourth communication is a request, which is received by the user, to perform the operation and further by determining whether the request corresponds to a business role of the user.

### Claim 16 (depends on 10)

The method of claim 10, further comprising:
generating embeddings that represent communications in the communication history of the user;
wherein causing the AI model to generate the third security analysis result comprises:
causing the AI model to compare the embeddings, which represent the communications in the communication history of the user, to a reference embedding, which represents the explanation criterion; and
wherein an extent to which an embedding of the fourth communication corresponds to the reference embedding indicates whether the fourth communication provides the explanation of the purpose of the operation that satisfies the explanation criterion.

### Claim 17 (depends on 10)

The method of claim 10, wherein the first security analysis result comprises a first security score, which is a numerical representation of whether the communication history of the user comprises at least one of the first communication that initiates the interaction with another user in the organization or the second communication that is addressed specifically to the user from another user in the organization;
wherein the second security analysis result comprises a second security score, which is a numerical representation of whether the communication history of the user comprises the third communication from the user that references the operation;
wherein the third security analysis result comprises a third security score, which is a numerical representation of whether the communication history of the user comprises the fourth communication that provides the explanation of the purpose of the operation that satisfies the explanation criterion;
wherein the method further comprises:
generating a combined security score by combining the first, second, and third security scores; and
wherein triggering the execution of the instruction comprises:
causing the security action to be performed with regard to the operation as a result of the combined security score satisfying the security criterion.

### Claim 18 (depends on 10)

The method of claim 10, wherein generating the first security analysis result comprises:
generating a first security score, which corresponds to at least one of a number of first communications in the communication history that are from the user and that initiate an interaction with another user in the organization or a number of second communications in the communication history that are addressed specifically to the user from another user in the organization;
wherein generating the second security analysis result comprises:
generating a second security score, which corresponds to a number of third communications in the communication history that are from the user and that reference the operation;
wherein causing the AI model to generate the third security analysis result comprises:
causing the AI model to generate a third security score, which corresponds to an extent to which the explanation provided by the fourth communication satisfies the explanation criterion; and
wherein performing the security action comprises:
performing the security action with regard to the operation as a result of a combination of the first, second, and third security scores satisfying the security criterion.

### Claim 19

A computer program product comprising a computer-readable storage medium having instructions recorded thereon for enabling a processor-based system to perform operations, the operations comprising:
generating a first security analysis result by determining whether a communication history of a user in an organization who performs an operation that triggers a security event comprises at least one of a first communication from the user that initiates an interaction with another user in the organization or a second communication that is addressed specifically to the user from another user in the organization;
generating a second security analysis result by determining whether the communication history of the user comprises a third communication from the user that references the operation;
generating a third security analysis result, which comprises determining whether the communication history of the user comprises a fourth communication that provides an explanation of a purpose of the operation that satisfies an explanation criterion, using an artificial intelligence (AI) model by causing the AI model to analyze content of the fourth communication in context of the explanation criterion; and
in response to the security event, triggering execution of an instruction, which causes a security action to be performed with regard to the operation as a result of a combination of the first, second, and third security analysis results satisfying a security criterion.

### Claim 20 (depends on 19)

The computer program product of claim 19, wherein the third security analysis result is generated by performing at least one of the following:
determining whether a value of an attribute of a user-generated communication, which is generated by the user and which references the operation, in the communication history of the user corresponds to a first reference value of the attribute of a reference user-generated communication, which references a first historical operation that corresponds to the operation performed by the user;
determining whether a value of an attribute of a system-generated communication, which is generated by a system utilized by the user and which references the operation, corresponds to a second reference value of the attribute of a reference system-generated communication, which references a second historical operation that corresponds to the operation performed by the user; or
determining whether the fourth communication is a request, which is received by the user, to perform the operation and further by determining whether the request corresponds to a business role of the user.
