Non-Invertible Deceptive Infusion of Data (DIOD) Methodology for Critical Data Communication

Back to all technologies
Download as PDF
Due to the potential threat of data reverse-engineering to find sensitive information, owners of critical infrastructure that seek to leverage the benefits of AI to automate functions and processes are often reluctant to share information with third parties. Now, researchers at Purdue University have created a novel scalable and efficient data masking paradigm for the secure exchange of data pertaining to critical infrastructure. The approach obfuscates the identity and critical data of the system through an innovative mathematical algorithm that transforms the data in a manner that preserves the information relevant to AI and misleads unscrupulous third parties attempting to reverse-engineer the data to an entirely different benign system. The approach to the problem represents a paradigm shift for the field and contrasts with existing data masking tools such as encryption and substitution which often render the data unusable for AI.

- Allows for critical data to be given to a third party (ex. contractor, customer, etc.) without compromising details regarding the source system
- Doesn't impact the application of AI on the final dataset
- Addresses several data sharing legal and export concerns
- Facilitates seamless flow of data between data owners and users

- Critical data communication
- Data sharing
- Data privacy
- Government/Corporate Intelligence
Jul 30, 2021
United States
Purdue Office of Technology Commercialization
The Convergence Center
101 Foundry Drive, Suite 2500
West Lafayette, IN 47906

Phone: (765) 588-3475
Fax: (765) 463-3486