Back to results
US 2026/0310698 A1Published application
Systems and methods for automated prompt and response moderation through generalized autonomous intelligent agents
AssigneeAbysius Inc.
InventorKevin Friesth
Life of this patent
Estimating the term…
Abstract
Systems and methods provide a framework for the automated evaluation of user prompts to identify impermissible intents and, in response to identifying an impermissible intent, performing any required remedial actions. Further, these systems and methods provide for automated evaluation of responses generated through domain-specific artificial intelligence (AI) models and according to applicable rules and standards to modify these responses prior to dissemination to users.
Claims21
A computer-implemented method comprising:
- detecting a user prompt associated with a user, wherein the user prompt is received during an ongoing communications session between a user and a generalized autonomous intelligent agent;
- processing the user prompt through the generalized autonomous intelligent agent to detect an intent;
- identifying a set of applicable rules corresponding to different intents, wherein the set of applicable rules defines one or more impermissible intents and one or more ethical limitations associated with a set of user characteristics corresponding to the user;
- transmitting the user prompt to a set of domain-specific artificial intelligence models to solicit a response to the user prompt, wherein the user prompt is transmitted as a result of the intent not corresponding to the one or more impermissible intents, and wherein when the user prompt is received by the set of domain-specific artificial intelligence models, the set of domain-specific artificial intelligence models collaboratively processes the user prompt to generate a set of responses;
- modifying the set of responses according to the set of applicable rules to generate a tailored response, wherein the tailored response is devoid of content corresponding to the one or more ethical limitations;
- and
- updating an interface associated with the ongoing communications session to provide the tailored response.
4 more elements
A system, comprising:
- one or more processors;
- and
- memory storing thereon instructions that, as a result of being executed by the one or more processors, cause the system to:
- detect a user prompt associated with a user, wherein the user prompt is received during an ongoing communications session between a user and a generalized autonomous intelligent agent;
- process the user prompt through the generalized autonomous intelligent agent to detect an intent;
- identify a set of applicable rules corresponding to different intents, wherein the set of applicable rules defines one or more impermissible intents and one or more ethical limitations associated with a set of user characteristics corresponding to the user;
- transmit the user prompt to a set of domain-specific artificial intelligence models to solicit a response to the user prompt, wherein the user prompt is transmitted as a result of the intent not corresponding to the one or more impermissible intents, and wherein when the user prompt is received by the set of domain-specific artificial intelligence models, the set of domain-specific artificial intelligence models collaboratively processes the user prompt to generate a set of responses;
- modify the set of responses according to the set of applicable rules to generate a tailored response, wherein the tailored response is devoid of content corresponding to the one or more ethical limitations;
- and
- update an interface associated with the ongoing communications session to provide the tailored response.
7 more elements
A non-transitory computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to:
- detect a user prompt associated with a user, wherein the user prompt is received during an ongoing communications session between a user and a generalized autonomous intelligent agent;
- process the user prompt through the generalized autonomous intelligent agent to detect an intent;
- identify a set of applicable rules corresponding to different intents, wherein the set of applicable rules defines one or more impermissible intents and one or more ethical limitations associated with a set of user characteristics corresponding to the user;
- transmit the user prompt to a set of domain-specific artificial intelligence models to solicit a response to the user prompt, wherein the user prompt is transmitted as a result of the intent not corresponding to the one or more impermissible intents, and wherein when the user prompt is received by the set of domain-specific artificial intelligence models, the set of domain-specific artificial intelligence models collaboratively processes the user prompt to generate a set of responses;
- modify the set of responses according to the set of applicable rules to generate a tailored response, wherein the tailored response is devoid of content corresponding to the one or more ethical limitations;
- and
- update an interface associated with the ongoing communications session to provide the tailored response.
4 more elements
Dependencies are inferred from the claim text.