TripleBlind Updates Virtual Private Data Sharing Solution to Enhance Performance and Simplify the User Experience
TripleBlind, the virtual private data sharing solution that accelerates the ability of enterprises to commercialize data while preserving privacy and enforcing compliance with all worldwide data privacy regulations, has updated and expanded its solution offering to address additional use cases.
With TripleBlind, financial institutions can share data to create a comprehensive view of consumers and create more effective anti-fraud and anti-money laundering strategies. Healthcare organizations can build more diverse patient data sets enabling development of highly accurate diagnostic algorithms, develop better treatments and drugs.
TripleBlind’s latest upgrade is built to allow data projects to progress as usual, but with an added layer of privacy. The main focus of the solution design is the user experience; the end user should never have to be an expert in the underlying cryptographic primitives in order to use the technology.
Since Python and R are the chosen languages for the majority of machine learning and data scientist users, TripleBlind’s Blind Data Tools and Blind AI Tools support these languages and the powerful data analysis and machine learning libraries made available through each. Using TripleBlind’s Blind Data Tools, the user is able to perform all the typical tasks involved in the data pre-processing pipeline, as well as perform tasks including logical aggregations, redactions, and Blind Sampling to ensure the quality of the data before usage.
TripleBlind’s Blind AI Tools support training a wide range of AI network architectures so that any desired architecture can be built on nearly any data type, without needing to “see” the data. TripleBlind users can also run Blind Inferences on trained models without ever physically exchanging the data or the model, allowing deployment of AI models across the globe.
With TripleBlind’s Blind Compute functionality, data and algorithms stay private and remain usable on automatically de-identified data. The company’s novel cryptographic method of splitting calculations in parts that can be run in parallel saves storage capacity and time, while further boosting privacy. Since TripleBlind’s APIs present themselves as similar to well-known frameworks such as Pandas, PyTorch, Scikit-Learn and Tensorflow, any number of them can be easily implemented and used by third parties. Only a few lines of code are needed to add privacy to an enterprise’s existing infrastructure.
With TripleBlind’s Blind Query, enterprises can bring together the data they need without privacy risk to them or their customers. Blind Query provides several ways to search, join and analyze data from disparate sources, while TripleBlind’s Blind Algorithm Tools enable the easy distribution of algorithms while maintaining strict security and theft prevention, enabling greater and safer collaboration.
For even greater accessibility, TripleBlind’s updated web user interface now allows enterprises to define and run a no-code drag-and-drop version of Blind Join. It logically aggregates two or more datasets for analysis in real time, without the user needing to look at a single line of code with all the same privacy benefits applied.
“Our Blind Virtual Exchange API is a simple code snippet that ensures raw data is never moved or exposed. Our API enables collaboration between entities, allowing operation on every detail of that data without ever transmitting unencrypted or non-aggregate information, while also enforcing permissions and providing an audit trail of use,” said Steve Penrod, Vice President of Product Development at TripleBlind. “Our Blind Algorithm Tools will empower financial and healthcare enterprises to easily distribute algorithms while maintaining strict security and theft prevention.”
[To share your insights with us, please write to email@example.com]