# US 12,182,237 B2 · Learning cognitive access control service > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,182,237 B2 - Document ID: US12182237B2 - Status: Granted - Kind: B2 - Application number: 17/540,031 - Filed: 2021-12-01 - Published: 2023-06-01 - Granted: 2024-12-31 - Earliest claimed: 2021-12-01 - Assignees: ServiceNow, Inc. - Applicants: ServiceNow, Inc. - Inventors: Ashton Mozano; Mohamed Sofiane Talmat - Web page: https://patentgrep.com/patent/US12182237B2 ## CPC - G06F 21/316: {by observing the pattern of computer usage, e.g. typical user behaviour} - G06F 21/604: {Tools and structures for managing or administering access control systems} - G06F 21/6218: {to a system of files or objects, e.g. local or distributed file system or database} - G06F 2221/2141: Access rights, e.g. capability lists, access control lists, access tables, access matrices - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract An indication associated with a request to access a protected object by a subject is received. Using one or more processors, application level behavioral patterns of the subject, context of the request by the subject, usage patterns associated with the protected object, and a current system state are automatically analyzed using one or more machine learning models to determine an analysis result associated with whether to grant the subject access to the protected object. An access control mechanism for the protected object is automatically modified based on the analysis result. ## Claims ### Claim 1 A method, comprising: receiving an indication associated with a request to access a protected object by a subject; using one or more processors to automatically analyze at least application level behavioral patterns of the subject, context of the request by the subject, usage patterns associated with the protected object, and a current system state using one or more machine learning models to determine an analysis result associated with whether to grant the subject access to the protected object, wherein the analysis result includes a plurality of predicted consequential effects of the access of the protected object determined using at least a portion of the one or more machine learning models; and automatically modifying an access control mechanism for the protected object based on the analysis result including the plurality of predicted consequential effects determined using at least a portion of the one or more machine learning models. ### Claim 2 (depends on 1) The method of claim 1, wherein the access control mechanism for the protected object is automatically modified for a limited amount of time. ### Claim 3 (depends on 1) The method of claim 1, wherein the access control mechanism utilizes an Identity Based Access Control (IBAC) and Access Control Lists (ACLs), Role Based Access Control (RBAC), Attribute Based Access Control (ABAC), or Next Generation Access Control (NGAC). ### Claim 4 (depends on 1) The method of claim 1, further comprising: evaluating a result of the modification to the access control mechanism; and retraining the one or more machine learning models based on the evaluated result. ### Claim 5 (depends on 1) The method of claim 1, wherein the modified access control mechanism is a conditional modification. ### Claim 6 (depends on 5) The method of claim 5, further comprising: identifying a failure to meet the conditional modification; and automatically revoking the automatic modification to the access control mechanism for the protected object. ### Claim 7 (depends on 1) The method of claim 1, wherein determining the analysis result associated with whether to grant the subject access to the protected object includes: identifying a prospective action to perform; evaluating a confidence rating corresponding to whether an intended consequence included the plurality of predicted consequential effects will occur; and in response to a determination that the confidence rating exceeds a threshold value, performing the prospective action. ### Claim 8 (depends on 1) The method of claim 1, wherein determining the analysis result associated with whether to grant the subject access to the protected object includes identifying a first, a second, a third, and a fourth order effect included the plurality of predicted consequential effects determined using at least a portion of the one or more machine learning models. ### Claim 9 (depends on 1) The method of claim 1, wherein determining the analysis result associated with whether to grant the subject access to the protected object includes creating a graphical representation of a current system context, wherein the current system context includes details extracted from the current system state and the context of the request by the subject. ### Claim 10 (depends on 1) The method of claim 1, wherein automatically modifying the access control mechanism for the protected object based on the analysis result includes providing a justification of the modification, wherein the justification is based on a system policy. ### Claim 11 A system, comprising: one or more processors; and a memory coupled to the one or more processors, wherein the memory is configured to provide the one or more processors with instructions which when executed cause the one or more processors to: receive an indication associated with a request to access a protected object by a subject; automatically analyze at least application level behavioral patterns of the subject, context of the request by the subject, usage patterns associated with the protected object, and a current system state using one or more machine learning models to determine an analysis result associated with whether to grant the subject access to the protected object, wherein the analysis result includes a plurality of predicted consequential effects of the access of the protected object determined using at least a portion of the one or more machine learning models; and automatically modify an access control mechanism for the protected object based on the analysis result including the plurality of predicted consequential effects determined using at least a portion of the one or more machine learning models. ### Claim 12 (depends on 11) The system of claim 11, wherein the access control mechanism for the protected object is automatically modified for a limited amount of time. ### Claim 13 (depends on 11) The system of claim 11, wherein the access control mechanism utilizes an Identity Based Access Control (IBAC) and Access Control Lists (ACLs), Role Based Access Control (RBAC), Attribute Based Access Control (ABAC), or Next Generation Access Control (NGAC). ### Claim 14 (depends on 11) The system of claim 11, wherein the memory is further configured to provide the one or more processors with instructions which when executed cause the one or more processors to: evaluate a result of the modification to the access control mechanism; and retrain the one or more machine learning models based on the evaluated result. ### Claim 15 (depends on 11) The system of claim 11, wherein the modified access control mechanism is a conditional modification, and wherein the memory is further configured to provide the one or more processors with instructions which when executed cause the one or more processors to: identify a failure to meet the conditional modification; and automatically revoke the automatic modification to the access control mechanism for the protected object. ### Claim 16 (depends on 11) The system of claim 11, wherein determining the analysis result associated with whether to grant the subject access to the protected object includes: identifying a prospective action to perform; evaluating a confidence rating corresponding to whether an intended consequence included the plurality of predicted consequential effects will occur; and in response to a determination that the confidence rating exceeds a threshold value, performing the prospective action. ### Claim 17 (depends on 11) The system of claim 11, wherein determining the analysis result associated with whether to grant the subject access to the protected object includes identifying a first, a second, a third, and a fourth order effect included the plurality of predicted consequential effects determined using at least a portion of the one or more machine learning models. ### Claim 18 (depends on 11) The system of claim 11, wherein determining the analysis result associated with whether to grant the subject access to the protected object includes creating a graphical representation of a current system context, wherein the current system context includes details extracted from the current system state and the context of the request by the subject. ### Claim 19 (depends on 11) The system of claim 11, wherein automatically modifying the access control mechanism for the protected object based on the analysis result includes providing a justification of the modification, wherein the justification is based on a system policy. ### Claim 20 A non-transitory computer readable storage medium comprising computer instructions for: receiving an indication associated with a request to access a protected object by a subject; automatically analyzing at least application level behavioral patterns of the subject, context of the request by the subject, usage patterns associated with the protected object, and a current system state using one or more machine learning models to determine an analysis result associated with whether to grant the subject access to the protected object, wherein the analysis result includes a plurality of predicted consequential effects of the access of the protected object determined using at least a portion of the one or more machine learning models; and automatically modifying an access control mechanism for the protected object based on the analysis result including the plurality of predicted consequential effects determined using at least a portion of the one or more machine learning models.