# US 2026/0087165 A1 · Secure AI Sandbox For Data Query Transformation And Refinement

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

- Number: US 2026/0087165 A1
- Document ID: US20260087165A1
- Status: Published application
- Kind: A1
- Application number: 19/334,075
- Filed: 2025-09-19
- Published: 2026-03-26
- Earliest claimed: 2024-09-20
- Inventors: Raghav Ramabadran; Saurav Kumar Singh
- Web page: https://patentgrep.com/patent/US20260087165A1

## CPC

- G06F 16/9035: Filtering based on additional data, e.g. user or group profiles
- G06F 21/6227: {where protection concerns the structure of data, e.g. records, types, queries}
- G06F 11/3409: {for performance assessment}
- G06F 16/2452: Query translation
- G06F 16/3329: Natural language query formulation
- G06F 16/33295: {in dialogue systems}
- G06F 16/3334: {Selection or weighting of terms from queries, including natural language queries}
- G06F 16/335: Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536)
- G06F 16/387: using geographical or spatial information, e.g. location
- G06F 21/6254: {by anonymising data, e.g. decorrelating personal data from the owner's identification}
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/045: Combinations of networks
- G06N 3/0455: Auto-encoder networks; Encoder-decoder networks
- G06N 3/047: Probabilistic or stochastic networks
- G06N 3/0475: Generative networks
- G06N 3/08: Learning methods
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- G06N 3/094: Adversarial learning
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A query associated with an entity is received. The query is transformed into a generic data request that excludes proprietary information. The generic data request is transmitted to at least one external artificial intelligence model. External data responsive to the generic data request is received. The external data is combined with proprietary data associated with the entity to generate a response, and the response is provided to the entity.

## Claims

### Claim 1

A method for processing data using artificial intelligence, comprising:
receiving a query associated with an entity;
transforming the query into a generic data request that excludes proprietary information;
transmitting the generic data request to at least one external artificial intelligence model;
receiving external data responsive to the generic data request;
combining the external data with proprietary data associated with the entity to generate a response; and
providing the response to the entity.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
determining that the query requires the external data; and
selecting the at least one external artificial intelligence model based on requirements of the generic data request.

### Claim 3 (depends on 1)

The method of claim 1, wherein the at least one external artificial intelligence model comprises a sequence of large language models orchestrated to process the generic data request.

### Claim 4 (depends on 1)

The method of claim 1, wherein combining the external data with the proprietary data comprises:
processing the external data using a custom language model implemented as a mixture-of-experts model; and
generating the response within a secure sandbox environment to maintain data isolation.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
training a custom artificial intelligence model on the proprietary data of the entity; and
using the custom artificial intelligence model to perform the combining of the external data with the proprietary data.

### Claim 6 (depends on 1)

The method of claim 1, wherein the transforming is executed within a secure environment that is isolated from external systems.

### Claim 7 (depends on 1)

The method of claim 1, wherein transforming the query comprises:
removing one or more of a client name, product identifier, confidential metric, or proprietary business strategy.

### Claim 8

A system, comprising:
a memory subsystem; and
processing circuitry, the processing circuitry configured to execute instructions stored in the memory subsystem to:
receive a query associated with an entity;
transform the query into a generic data request that excludes proprietary information;
transmit the generic data request to at least one external artificial intelligence model;
receive external data responsive to the generic data request;
combine the external data with proprietary data associated with the entity to generate a response; and
provide the response to the entity.

### Claim 9 (depends on 8)

The system of claim 8, wherein, to combine the external data with the proprietary data, the processing circuitry is configured to execute instructions stored in the memory subsystem to:
apply a custom artificial intelligence model trained on client-specific data to refine the external data.

### Claim 10 (depends on 8)

The system of claim 8, wherein, to receive the query, the processing circuitry is configured to execute instructions stored in the memory subsystem to:
receive the query via a problem-solving agent executing within a secure environment associated with the entity.

### Claim 11 (depends on 8)

The system of claim 8, wherein, to provide the response, the processing circuitry is configured to execute instructions stored in the memory subsystem to:
present the response via a user interface that indicates whether the response includes externally sourced data.

### Claim 12 (depends on 8)

The system of claim 8, wherein, to combine the external data with the proprietary data, the processing circuitry is configured to execute instructions stored in the memory subsystem to:
perform a post-generative refinement (PGR) process in which content from external models is edited based on entity-specific preferences.

### Claim 13 (depends on 8)

The system of claim 8, wherein, to transmit the generic data request, the processing circuitry is configured to execute instructions stored in the memory subsystem to:
include prompt context describing a content generation or analysis goal while omitting entity-specific instructions.

### Claim 14 (depends on 8)

The system of claim 8, wherein, to transmit the generic data request to at least one external artificial intelligence model, the processing circuitry is configured to:
transmit to a sequence of two or more external artificial intelligence models.

### Claim 15 (depends on 8)

The system of claim 8, wherein combining the external data with the proprietary data is performed within a secure sandbox environment that is a dedicated, isolated computing instance that prevents commingling of data between different entities.

### Claim 16 (depends on 8)

The system of claim 8, the processing circuitry further configured to:
determine a processing sequence of the at least one external artificial intelligence model based on the generic data request.

### Claim 17

One or more non-transitory computer readable storage media comprising instructions that, when executed by one or more processors, perform operations comprising:
receiving a query associated with an entity;
transforming the query into a generic data request that excludes proprietary information;
transmitting the generic data request to at least one external artificial intelligence model;
receiving external data responsive to the generic data request;
combining the external data with proprietary data associated with the entity to generate a response; and
providing the response to the entity.

### Claim 18 (depends on 17)

The one or more non-transitory computer readable storage media of claim 17, wherein providing the response to the entity includes delivering an analytical document that combines external market intelligence with internal proprietary insights.

### Claim 19 (depends on 17)

The one or more non-transitory computer readable storage media of claim 17, wherein the external data responsive to the generic data request includes publicly available market analysis, regulatory approval timelines, or pricing strategies.

### Claim 20 (depends on 17)

The one or more non-transitory computer readable storage media of claim 17, wherein transforming the query comprises prompting a language model to use semantic abstraction techniques to map client-specific queries to predefined generic formats while preserving analytical requirements.
