AI-Generated Authentication Queries Against Phishing and Keyloggers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional type-1 authentication methods are vulnerable to password guessing, phishing attacks, and keylogger malware, and password managers are susceptible to security breaches, compromising user data integrity.
Innovation Solution
Implement AI-generated natural language queries and machine learning models to authenticate users based on personalized information, eliminating the need for remembering passwords and enhancing security through dynamic, impervious authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If type-1 authentication uses passwords, then users can access accounts, but the system becomes vulnerable to guessing, phishing, and keylogger malware
Solution Approach 1:
The patent replaces the mechanical/password-based authentication system with an AI-driven natural language processing system. Instead of relying on static passwords that can be guessed or stolen, the system uses machine learning models to analyze user responses to dynamically generated questions, substituting the traditional credential-based mechanism with an intelligent assessment approach that adapts to each authentication attempt.
Solution Approach 2:
The authentication system transitions from static passwords to dynamic AI-generated questions. The system continuously adapts the authentication process by generating new natural language questions based on user profiles and context, making each authentication attempt unique and resistant to replay attacks. This dynamic approach ensures that even if an attacker obtains a response, it cannot be reused for subsequent authentication.
2Ease of operation
If users use the same password across many accounts, then ease of operation improves, but security deteriorates due to susceptibility to security breaches
Solution Approach 1:
The patent extracts the authentication burden from the user by eliminating the need to remember and manage multiple passwords. Instead of requiring users to maintain secure but complex passwords across numerous accounts, the system extracts the authentication function and performs it automatically through AI-based verification, freeing users from the security management task while maintaining high security levels.
Solution Approach 2:
The AI authentication system acts as an intermediary between the user and the authentication mechanism. Rather than users directly managing credentials, the intermediary system handles passwordless verification by analyzing natural language responses, thereby removing the security risks associated with password reuse while maintaining convenient access.
3Device complexity
If type-1 authentication is used, then implementation is simple, but it is vulnerable to phishing attacks where bad actors impersonate trustworthy websites
Solution Approach 1:
The system incorporates feedback mechanisms where the AI model continuously analyzes user responses to generated questions and adjusts the authentication process accordingly. This feedback loop enables the system to detect anomalies and adapt to phishing attempts in real-time, making it resistant to impersonation attacks while maintaining implementation simplicity through automated verification.
Solution Approach 2:
The authentication system performs preliminary actions by pre-generating and storing a diverse set of natural language questions associated with user profiles before actual authentication occurs. This preliminary preparation enables the system to rapidly respond to authentication requests with appropriate questions, making it difficult for phishers to predict or replicate the authentication flow.
4Ease of operation
If passwords are used for authentication, then access can be granted, but keylogger malware can record all keystrokes including passwords
Solution Approach 1:
The patent fundamentally substitutes the mechanical keystroke-based authentication with a voice- or text-based natural language interaction system. By replacing the mechanical input method with AI-driven understanding of natural language responses, the system eliminates the vulnerability to keyloggers that target keystrokes, while maintaining the ability to verify user identity through intelligent question-response pairs.
Data Source
AI summary
A method and electronic device for type-1 authentication based on AI-generated queries is provided. The method includes receiving, via a user device, a request to access a user account and extracting, based on the request, a NL query of a set of NL queries that is generated based on information associated with a personal life of the user. The method further includes controlling the user device to render the NL query and receiving a response to the NL query from the user device. The method further includes determining a weighted score for the response based on application of a ML model on the response. The weighted score indicates a level of difficulty of the NL query and an accuracy of the response with respect to a NL response for the NL query. The method further includes authenticating the request if the weighted score is above a threshold score.