# US 2026/0075426 A1 · Multi-model artificial intelligence-based scam checker

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

- Number: US 2026/0075426 A1
- Document ID: US20260075426A1
- Status: Published application
- Kind: A1
- Application number: 19/285,990
- Filed: 2025-07-30
- Published: 2026-03-12
- Earliest claimed: 2024-09-10
- Assignees: Trend Micro Incorporated
- Applicants: Trend Micro Incorporated
- Inventors: Yi-Fen CHEN; Yen-Ying HUANG; Jing-Lin WU; Ming-Cheng PAN
- Web page: https://patentgrep.com/patent/US20260075426A1

## CPC

- H04W 12/35: {Protecting application or service provisioning, e.g. securing SIM application provisioning}
- H04W 12/122: Counter-measures against attacks; Protection against rogue devices
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)

## Abstract

Disclosed is a scam checker for checking contents for scams. The scam checker includes several artificial intelligence (AI) models to decompose an input content, generate a plan for collecting evidence for determining whether the input content is indicative of a scam, collect the evidence, and evaluate the collected evidence to generate a summary that indicates whether the input content is indicative of a scam.

## Claims

### Claim 1

A method of checking an input content for scams, the method to be performed by a scam checker that comprises a plurality of artificial intelligence (AI) models, the method comprising:
receiving the input content from a user device;
decomposing, by one of the plurality of AI models, the input content to one or more constituent data modalities of the input content;
generating, by one of the plurality of AI models, an evidence gathering plan based at least on the one or more constituent data modalities of the input content;
collecting, by one of the plurality of AI models, evidence based at least on the evidence gathering plan; and
evaluating, by one of the plurality of AI models, the collected evidence to generate a summary, wherein the summary includes a conclusion as to whether the input content is indicative of a scam.

### Claim 2 (depends on 1)

The method of claim 1, wherein the summary includes a reasoning behind the conclusion.

### Claim 3 (depends on 1)

The method of claim 1, wherein the user device is a smartphone, the input content is a screenshot of an app running on the smartphone, and the one or more constituent data modalities include an image data of the screenshot and a text of the screenshot.

### Claim 4 (depends on 3)

The method of claim 3, wherein the evidence gathering plan includes checking a reputation service for a reputation of a poster the posted an image of the image data on a social media site.

### Claim 5 (depends on 3)

The method of claim 3, wherein the evidence gathering plan includes searching the public Internet for an image of the image data.

### Claim 6 (depends on 1)

The method of claim 1, wherein the user device is a smartphone, the input content is a text message, and the one or more constituent data modalities include a text of the text message.

### Claim 7

A method of checking an input content for scams, the method comprising:
receiving a screenshot of a message on a smartphone employed by a user;
decomposing the screenshot into text;
evaluating the text, using an artificial intelligence (AI) model, for scams; and
alerting the user responsive to detecting that the input content is indicative of a scam.

### Claim 8 (depends on 7)

The method of claim 7, further comprising:
decomposing the screenshot into image data; and
evaluating the image data, using an AI model, for scams.

### Claim 9 (depends on 7)

The method of claim 7, wherein alerting the user includes generating a summary that includes a conclusion that the message is indicative of the scam and a reasoning behind the conclusion.

### Claim 10

A method of checking contents for scams, the method to be performed by a scam checker that comprises a plurality of artificial intelligence (AI) models, the method comprising:
receiving a screenshot of an app of a smartphone employed by a user;
extracting, by one of the plurality of AI models, text from the screenshot;
generating, by one of the plurality of AI models, an evidence gathering plan based at least on the text;
collecting, by one of the plurality of AI models, evidence based at least on the evidence gathering plan;
evaluating, by one of the plurality of AI models, the collected evidence to generate a summary that indicates whether the screenshot includes information that is indicative of a scam; and
providing the summary to the user.

### Claim 11 (depends on 10)

The method of claim 10, wherein the summary includes a conclusion that the text is indicative of the scam and a reasoning behind the conclusion.

### Claim 12 (depends on 10)

The method of claim 10, wherein the evidence gathering plan includes checking a reputation service for a reputation of a poster that posted an image of the image data on a social media site.

### Claim 13 (depends on 10)

The method of claim 10, further comprising:
decomposing the screenshot into image data,
wherein the evidence gathering plan includes searching the public Internet for an image of the image data.
