Abstract: A system for securing a large language model (LLM) service against LLM privacy attacks. The system may comprise a processor that executes instructions that cause the processor to: interface with each output of the LLM service and each client network of the LLM service; monitor each output of the LLM to detect at least one client query textual response; detect and redact the at least one client query textual response according to an evaluation of that response and at least one current client privilege level that is assigned to at least one target client network account to which the at least one client query textual response is directed; and transmit a result of the redacting to a client device of the at least one target client network account.

Abstract: A method for facilitating resource management by using predictive analytics is disclosed. The method includes aggregating, via an application programming interface, data from various sources, the data including end user data, resource data, and influential factor data; generating data products based on the aggregated data, the data products including a structured data set, an application, and a tool; training a first model by using the generated data products; determining predictive outputs by using the trained first model and the generated data products, each of the predictive outputs corresponding to a recommended action for management of resources; and publishing the predictive outputs to a downstream application.

Abstract: In some aspects, the techniques described herein relate to a method including: storing, in an application management database, an audit group, wherein the audit group identifies one or more computer applications; querying the application management database, wherein the querying returns a plurality of access control descriptions from the application management database; processing keywords from the plurality of access control descriptions with a machine learning (ML) model to determine a plurality of duty groups; associating access control permissions with the plurality of duty groups based on the keywords; and determining toxic combinations of the access control permissions based on database relationships between the plurality of duty groups, the audit group, and a user identifier.