A system (10) to evaluate one or more risks associated with an artificial intelligence platform and governance is provided The processing subsystem (50) includes a user input module (90) configured to receive input prompts from a user. The processing subsystem includes a processing module (100) to convert the input prompts into vectors. The processing subsystem includes a context evaluation module (110) to evaluate the artificial intelligence model adversarial attack signatures to generate an ethical attack score, security attack score and privacy attack score. The processing subsystem includes a summarization module to evaluate a probability score based on the ethical attack score, the security attack score and the privacy attack score. The summarization module is to identify an attack based on the probability score. The processing subsystem includes a flagging module to render the attack identified in a user interface. The processing subsystem includes a human feedback module to communicate feedbacks.
H04L 51/02 - User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
2.
SYSTEM AND METHOD TO ALTER QUERIES PROVIDED TO AN ARTIFICIAL INTELLIGENCE PLATFORM FOR PRESERVING PRIVACY
A system to alter queries provided to an artificial intelligence platform for preserving privacy is disclosed The system includes a query assessment module to identify sensitive attributes in prompts, a query annotation module to annotate them and a context and sensitivity based query paraphrasing module to generate synthetic attributes along with a privacy budget module to alter identified numerical values for creating synthetic prompt. The system includes a query output module to receive the synthetic privacy preserved query from upstream and forward it to downstream neural network with attention based language model and a query response origin module to receive an output generated by the artificial intelligence platform upon receiving the query. The system includes a query response assessment module to identify the sensitive synthetic attributes, a response privacy and context preservation module to modify the output to restore original entities and an output module to render the output.
H04L 51/02 - User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
3.
A SYSTEM FOR PRESCRIPTIVE AND PROTECTED PROMPT ENGINEERING AND A METHOD THEREOF
A system (100) for prescriptive and protected prompt engineering for a generative artificial intelligence is disclosed The system includes processing subsystem (108) which includes a user requirement input module (114) to receive one or more first queries prompt from a user, an interactive query module (116) suggests a plurality of second queries prompt to the user, a context gathering module (120) collects information related to a context of the input query, an objective gathering module (122) collect objectives of the first input queries, an assumption validation module (124) validates the selected option for the prompts, a prompt creation module (126), an expected utility validation module (128) validates the utility of the prompt, a risk assessment and mitigation module (130) understands risk of the input prompts and mitigation related to the prompts, and a prompt translation module (132) translates the selected prompt based on user's language preference.
H04L 51/02 - User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
4.
A SYSTEM FOR PRIVACY PROTECTED IDENTITY AND PROFILING PREVENTION AND A METHOD THEREOF
A system (100) for privacy protected identity and profiling prevention in a responsible artificial intelligence ecosystems is disclosed The system includes a user registration module (114) to register a user with a unique identity and authenticate the user, a generative artificial intelligence ecosystem integration module (116) receives sensitive information, provides privacy protected aggregate identity, accepts queries, connects the registered user with a plurality of downstream modules, performs a privacy threat modelling, comprehends a plurality of sensitive identities, and aggregates usage of a user's data, a profiling prevention module (120) identifies general sensitive information, annotate the identified information, a pseudo-identity mapping module (122) generates tokenized data and pseudonymized data for encryption and decrypting information, a secure key management module (124) generates a secure key for generating shares of the secure key, and a notification module (126) notifies the user regarding a database usage of the user by a third party.
The system (10) includes a processing subsystem including a context based risk treatment selection module to choose treatment based on potential adversarial attack patterns The processing subsystem includes a synthetic data based sensitive attribute modification module to replace identified sensitive attributes with a synthetic prompt. The processing subsystem includes a generative artificial intelligence-based input data optimization module to create a new prompt. The processing subsystem includes a treatment visualization module to help user visualize the attack mitigation. The processing subsystem includes a human in loop validation module to enable a user to understand the potentially adversarial risk present in the input data. The processing subsystem includes a residual risk acceptance module to enable validation of the identified adversarial attack and corresponding mitigation. The processing subsystem includes a consent capture module to enable organizations to capture the informed consent. The processing subsystem includes a recording module to store information. The processing subsystem includes a response display module to display the output.
A system (10) for a parental control in a generative artificial intelligence governance platform is disclosed The core modules include the parental requirement capture module to allow a parent to specify prerequisite conditions, harm identification, regulatory requirements, acceptable automated risk mitigation and alert services. A context based query assessment module to understand the context of a prompt, compare it against the configured parental requirement and quantify risk by interacting with subsequent questions. The query treatment module to identify risks associated with the prompt. Further, the query treatment module is configured to mitigate the risks from the child perspective based on a context of the intended action and parental requirement defined. The child prompt engineering module is configured to monitor the query, review the response and modify or block the response based on the policy defined. A parental alert module to alert the parent of the risk thereby ensuring parental control.
H04N 21/258 - Client or end-user data management, e.g. managing client capabilities, user preferences or demographics or processing of multiple end-users preferences to derive collaborative data
7.
A SYSTEM FOR MULTI MODAL AGGREGATION IN A GOVERNANCE PLATFORM AND A METHOD THEREOF
A system (100) for multi modal aggregation and governance platform is provided The system includes an identity protection module (114) providing a plurality of identities of a user at a plurality of downstream modules, convert them into a unified user identity and proxy the unified identity, a query receiving module (116) receives a plurality of queries from the user and classify them into one or more categories, a governance module (118) understands a context of the multimodal queries, control user prompt or chain of thought prompts, execute the multimodal queries via a responsible artificial intelligence model (120), control the queries or prompts, generate reports related to the risks involved in the input query and the response outputs, query and response distribution module (122) distributes executed the plurality of multimodal queries to the plurality of downstream generative AI modules (140), including translation and select a foremost response.
An Artificial Intelligence Governance system, ensuring responsible AI utilization is disclosed The system features a processing subsystem hosted on a server, orchestrating bidirectional communications across a network among numerous modules. A database module captures and manages diverse activities associated with interacting with Neural network with attention based AI models, while a user input module facilitates user engagement through API or UI. A configuration module defines organizational policies and regulatory requirements, ensuring responsible AI practices. An authentication module regulates user access based on roles, employing granular permission sets. A Neural network with attention based AI governance module uses a pre-trained AI model to classify risks in user prompts. A session management module establishes standardized API endpoints for seamless integration, and a dashboard module offers a summarized overview, governing responsible AI usage within the organization. This system provides transparency, and ethical AI practices across various organizational contexts.
A system (10) for input risk identification within an artificial intelligence responsible governance platform is disclosed The system includes a processing subsystem (50) hosted on a server (60), the system enables bidirectional communications among multiple modules. The core component, a neural network with attention based AI risk classification module (70), employs various submodules including privacy, financial, health, intellectual property, technology, and confidential information, each meticulously trained and configured. Leveraging advanced machine learning techniques, such as natural language processing, pattern recognition, and contextual analysis, these submodules identify and categorize risks within user input or information flows. The annotation module (140) ensures clarity by annotating identified risks, while the feedback module (150) collects user insights for continuous improvement. The output module (160) communicates risk classification data to downstream risk mitigation modules, offering a comprehensive solution for managing and mitigating diverse risks associated with artificial intelligence interactions in accordance with organizational policies and regulatory requirements.
A system (10) to predict, summarize and recommend risks in artificial intelligence responsible governance platform is disclosed The core module includes the risk analysis module to analyze input prompts using a Neural network with attention based artificial intelligence model. The risk analysis module further includes a domain level prediction module to predict the nature, purpose, scope and context of the input prompts, a time context prediction module to predict the time related aspects, a sensitivity prediction module to recognize and filter the occurrence of sensitive information, a likelihood and severity prediction module to evaluate and grade risks and its probability, a risk prediction module to finalize representation of selected risks, customize and share a predicted risk recommendation context to a user. Further, an output module configured to visualize summary of the identified risk, transfer the summary, recommended contextual risks to a downstream risk mitigation module.
H04L 51/02 - User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
patents
