Authenticator Detecting Deepfake Requests via Smart Device Media
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current authentication methods are inadequate in detecting fraudulent audio and video samples, allowing unauthorized access to sensitive data and accounts due to the vulnerability of facial recognition and audio recognition processes to deepfake technologies.
Innovation Solution
An authentication system that utilizes an authenticator with a processor and memory to receive and analyze audio and video samples, activate smart devices, and compare samples from different devices to detect fraudulent requests by identifying inconsistencies in context, location, and light reflections, thereby verifying the authenticity of user requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If facial recognition and audio recognition processes are implemented, then authentication capability is improved, but vulnerability to deepfake technologies increases
Solution Approach 1:
The system performs preliminary actions by activating smart devices and capturing media samples before the authentication decision is made. This allows the system to establish a baseline of authentic user behavior and environment, which is then compared against the authentication request to detect deepfakes.
Solution Approach 2:
The patent introduces smart devices (smartphone, smartwatch, smart speaker) as intermediaries between the user and the authentication system. These devices capture media samples and environmental data, serving as a trusted intermediary that verifies the authenticity of the user before granting access, thereby reducing vulnerability to deepfake attacks.
2Measurement precision
If multiple smart devices and media samples are utilized, then detection accuracy is improved, but device complexity increases
Solution Approach 1:
The authentication system is segmented into multiple independent smart devices (smartphone, smartwatch, smart speaker), each performing specific functions such as capturing media samples, detecting light reflections, or verifying audio patterns. This segmentation allows the system to achieve high detection accuracy through multiple verification points while maintaining manageable complexity at each device level.
Solution Approach 2:
The smart devices serve multiple functions: they act as authentication requestors, media sample collectors, environmental sensors, and communication interfaces. This multi-functionality reduces the need for dedicated specialized components, thereby managing system complexity while maintaining high detection accuracy through versatile device utilization.
Data Source
AI summary
An authentication system for granting access to an account associated with a user. An authenticator of the authentication system including a processor and a memory, the authenticator configured to: receive a request for authentication that identifies the user; access an authentication account associated with the user, wherein the authentication account indicates a smart device associated with the user; activate a control parameter of the smart device, wherein the control parameter may cause an effect in a media sample from the smart device; receive the media sample from the smart device and determine if the effect corresponding to the control parameter is observed in the media sample; and in response to a determination that the effect is not observed in the media sample, indicate that the request for authentication is a fraudulent request.