OneTrust, LLC

Methods, systems, and non-transitory computer readable storage media are disclosed for controlling access to artificial intelligence systems based on determined intent of queries. The disclosed system utilizes one or more digital content analysis models to determine an intent of one or more queries to an artificial intelligence system. The disclosed system utilizes the one or more digital content analysis models to determine an intended use of the artificial intelligence system. Additionally, the disclosed system determines whether the intent of the one or more queries aligns with the intended use of the artificial intelligence system by generating a similarity score and comparing the similarity score to a similarity threshold. Based on whether the intent aligns with the intended use, the disclosed system executes computing instructions to enable or block the one or more queries from being processed by the artificial intelligence system.

Methods, systems, and non-transitory computer readable storage media are disclosed for controlling access to artificial intelligence systems based on determined intent of queries. The disclosed system utilizes one or more digital content analysis models to determine an intent of one or more queries to an artificial intelligence system. The disclosed system utilizes the one or more digital content analysis models to determine an intended use of the artificial intelligence system. Additionally, the disclosed system determines whether the intent of the one or more queries aligns with the intended use of the artificial intelligence system by generating a similarity score and comparing the similarity score to a similarity threshold. Based on whether the intent aligns with the intended use, the disclosed system executes computing instructions to enable or block the one or more queries from being processed by the artificial intelligence system.

Aspects of the present invention provide methods, apparatuses, systems, computing devices, computing entities, and/or the like for detecting prejudice bias in machine-learning models and/or data sets used in training, testing, and/or validating the models. In accordance various aspects, a method is provided comprising: receiving a data set used for training, testing, and/or validating a model that comprises data instances; generating, using a classification model, a prediction of applicability for each sub-category of a plurality of sub-categories for each bias category of a plurality of bias categories for each data instance; determining that a particular sub-category for a particular bias category is applicable to a proportion of the data set, wherein predictions of applicability for the particular sub-category generated for the proportion of the data set satisfies a threshold; and determining, based on the proportion, that the data set has a prejudice bias with respect to the particular bias category.

Methods, systems, and non-transitory computer readable storage media are disclosed for facilitating modification of components of artificial intelligence computing applications via aggregated risk scores of individual artificial intelligence computing application components. The disclosed system generates data objects representing the artificial intelligence computing application components. The disclosed system determines mappings between the data objects based on relationships of the artificial intelligence computing application components. Furthermore, the disclosed system generates risk scores for the data objects representing the components of the artificial intelligence computing application by administering risk assessments that correspond to a system requirements framework. The disclosed system also generates an interactive aggregated risk indicator indicating contributions of the components to the artificial intelligence computing application in relation to the system requirements framework by combining the risk scores of the data objects according to the mappings between the data objects.

Aspects of the present invention provide methods, apparatuses, systems, computing devices, computing entities, and/or the like for detecting prejudice bias in machine-learning models and/or data sets used in training, testing, and/or validating the models. In accordance various aspects, a method is provided comprising: receiving a data set used for training, testing, and/or validating a model that comprises data instances; generating, using a classification model, a prediction of applicability for each sub-category of a plurality of sub-categories for each bias category of a plurality of bias categories for each data instance; determining that a particular sub-category for a particular bias category is applicable to a proportion of the data set, wherein predictions of applicability for the particular sub-category generated for the proportion of the data set satisfies a threshold; and determining, based on the proportion, that the data set has a prejudice bias with respect to the particular bias category.

Methods, systems, and non-transitory computer readable storage media are disclosed for managing implementation of machine-learning models within computing environments according to system requirements frameworks via common data objects. The disclosed system generates a common data object to represent an implementation of a machine-learning model with a data process. For example, the disclosed system determines attribute values of the common data object according to data objects representing the machine-learning model and related datasets. Furthermore, the disclosed system utilizes the common data object to validate the machine-learning model according to a digital representation of a system requirements framework that includes usage requirements for machine-learning models to store, process, transmit, or otherwise handle specific data types in specific ways for the one or more data processes within a computing environment. The disclosed systems also perform operations to implement, suspend, or otherwise modify the machine-learning model or datasets based on the validation.

Aspects of the present invention provide methods, apparatuses, systems, computing devices, computing entities, and/or the like for detecting prejudice bias in machine-learning models and/or data sets used in training, testing, and/or validating the models. In accordance various aspects, a method is provided comprising: receiving a data set used for training, testing, and/or validating a model that comprises data instances; generating, using a classification model, a prediction of applicability for each sub-category of a plurality of sub-categories for each bias category of a plurality of bias categories for each data instance; determining that a particular sub-category for a particular bias category is applicable to a proportion of the data set, wherein predictions of applicability for the particular sub-category generated for the proportion of the data set satisfies a threshold; and determining, based on the proportion, that the data set has a prejudice bias with respect to the particular bias category.