# US 2025/0148026 A1 · Systems and methods for independent audit and assessment framework for ai systems

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0148026 A1
- Document ID: US20250148026A1
- Status: Published application
- Kind: A1
- Application number: 18/942,117
- Filed: 2024-11-08
- Published: 2025-05-08
- Earliest claimed: 2023-11-08
- Assignees: Arizona Board of Regents on behalf of Arizona State University
- Inventors: Siddharth Srivastava; Pulkit Verma
- Web page: https://patentgrep.com/patent/US20250148026A1

## CPC

- G06F 16/953: Querying, e.g. by the use of web search engines
- G06F 16/243: {Natural language query formulation}
- G06F 16/29: Geographical information databases
- G06F 40/30: Semantic analysis
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Example systems and methods for AI capability assessment include query-based assessment of sequential decision making agents (SDMAs) in stochastic settings with minimal assumptions on SDMA internals. In these examples, a new approach is presented for modeling the capabilities of black-box AI systems by using an active learning approach that can effectively interact with the black-box AI systems and learn an interpretable probabilistic model describing the capabilities of the black-box AI systems

## Claims

### Claim 1

A method for autonomous assessment of sequential decision-making systems in stochastic settings, comprising:
accessing an input A, that is the software and/or integrated hardware constituting an agent capable of sequential decision making, or planning, and a simulator Sim that the agent A can interact with to execute its behavior;
inducing a system that accepts users' specification of objectives and computes and executes the behavior necessary for achieving them.
learning a model defining capabilities of the SDM system using query-based autonomous capability estimation, including:
defining M as the space of capability models expressible in terms of the set of interpretable concepts or predicates known to the user,
discovering and learning the true capabilities of the agent A as a model in M in a manner that clarifies to the user the A's capabilities and the effects of executing them, by posing queries to the agent of the SDM system and utilizing responses to the queries as data to eliminate inconsistent models from the set of capability models M, wherein the model as learned models the set of A's capabilities in stochastic settings.

### Claim 2 (depends on 1)

The method of claim 1, further comprising measuring a variational distance between the model and a true transition model associated with the SDMA system to evaluate how close the model is to the true transition model.

### Claim 3 (depends on 1)

The method of claim 1, further comprising using object-centric predicate representation including an abstraction of environment states to high-level logical states expressible in predicate vocabulary to induce the set of transition models.

### Claim 4 (depends on 1)

The method of claim 1, wherein the model as learned is a probabilistic model expressed in terms of the set of predicates and the set of capabilities.

### Claim 5 (depends on 1)

The method of claim 1, wherein the model as learned is a non-deterministic model expressed in terms of the set of predicates and the set of capabilities.

### Claim 6 (depends on 1)

The method of claim 1, wherein the set of predicates are first order logic predicates and the evaluation functions include Boolean evaluation functions.

### Claim 7 (depends on 1)

The method of claim 1, wherein the model is learned using input concept vocabulary modeled as binary-valued predicates.

### Claim 8 (depends on 1)

The method of claim 1, wherein the model is learned by utilizing syntax and semantics of known SDM model representation language and probabilistic planning definition language.
