Abstract
A method and system for artificial intelligence (AI) automation are disclosed. An AI model is used to determine a confidence level that an AI action on an object, such as AI document processing, meets a pre-defined target of processing accuracy. If the confidence level is above a confidence threshold, the AI action is adopted; if not, a manual action may be performed instead. Upon automatically processing N documents, a sub-population M of N documents is verified. A Monte Carlo simulation is usable to provide an overall confidence threshold including a statistical estimation of the accuracy of non-verified AI actions.
Full Text
What is claimed is:
A method and system for artificial intelligence (AI) automation are disclosed. An AI model is used to determine a confidence level that an AI action on an object, such as AI document processing, meets a pre-defined target of processing accuracy. If the confidence level is above a confidence threshold, the AI action is adopted; if not, a manual action may be performed instead. Upon automatically processing N documents, a sub-population M of N documents is verified. A Monte Carlo simulation is usable to provide an overall confidence threshold including a statistical estimation of the accuracy of non-verified AI actions.
Timeline
Filed
04/21/2026Published
08/20/2026Granted
Not AvailableIPC Codes(3)
G06V 30/19:Recognition using electronic means
G06F 30/20:Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F 30/30)
G06F 111/08:Probabilistic or stochastic CAD