team_board_blur.png
UN_Flags.png
trading_dashboard.png
Credit.png
Family.jpg
team_board_blur.png

Case Studies


Data privacy and security shouldn't be an afterthought.

SCROLL DOWN

Case Studies


Data privacy and security shouldn't be an afterthought.

The Big Data Privacy and Security Challenge

Everyone is trying to take advantage of their own (and others') data with advanced analytics and building machine learning models for more accurate predictions.  This goal is challenging enough for data science and IT teams; however, data security and privacy challenges across departments, companies and myriad regulatory landscapes make it a truly daunting endeavor.  

Three Targets

We are focused on solving the problem relating to these three areas, while enabling you to keep the data where it is or securely migrate to zero-knowledge cloud computing:

+ Running privacy-compliant analytics and machine learning models on your organization's sensitive data across departments, jurisdictions and regulatory bodies.

+ Building more robust and predictive models with alternative and private third party data sources without revealing any proprietary or sensitive data.

+ Monetizing the insight of your data without giving away your data.

data_jurisdiction.png
UN_Flags.png

Commercial Banking


Commercial and Private Banking

Commercial Banking


Commercial and Private Banking

Business Challenge

To conduct privacy-compliant analytics and machine learning across customers and jurisdictions that meet current and upcoming regulations such as GDPR and the UK Data Protection Bill, while opening the opportunity for secure cloud computing. Applications for zero-knowledge computing in Commercial, Retail and Private Banking include:

  • Customer marketing and portfolio evaluation
  • Fraud detection and AML investigation
  • Credit risk analysis
  • Secure back-office operations

Solution

As reported in the Wall Street Journal, customers like ING Belgium use the XOR Secret Computing™ Engine to build analytical models using their data from multiple countries with stringent data security and personal privacy rules, like Switzerland and Luxembourg.  Proprietary algorithms generated by ING data science teams are compiled with XOR and secretly computed by all regional datacenters and/or cloud services providers without revealing any sensitive information; no PII is exported from any jurisdiction.  With more data sources, the teams can build improved models and more accurately predict anything from loan defaults and payment fraud to identifying which products or services a customer is likely to buy.

 

trading_dashboard.png

InvestmentBanking


Investment Banks and Hedge Funds

InvestmentBanking


Investment Banks and Hedge Funds

Business Challenge

There is a well documented paradox in trading alpha for quantitative strategies; the buyer wants to evaluate the data to see how it can improve their models before purchasing, while the seller doesn't want to give away their IP without being paid.  Traditional methods of providing sample sets or older historical data can lead to over-fitting and poor prediction when applied to new data streams.  How can correlation and causality be calculated without revealing the inputs?  How do you monetize insights from, or even assign a value to, proprietary trading information without giving away the data, particularly under MiFID II?

Solution

Investment banks, hedge funds and third party data providers leverage Secret Computing™ capabilities to efficiently calculate the impact of private data sources on existing models to determine its value, without exposing the data. The XOR Engine provides arithmetic precision that yields the same results as plaintext calculations, even with complex machine learning algorithms running on multiple massive datasets. Computation can be on-premise or secretly shared to cloud providers for zero-knowledge analytics, with the results securely delivered only to predefined recipients.  

Credit.png

Credit


Credit and Payments

Credit


Credit and Payments

Business Challenge

The analytics team wants to do basic statistics and train their supervised learning models with sensitive customer data on their network (or cloud provider), but they don't have security permissions to read the data.  Attempts to anonymize the data removes valuable features in their model, and it is well-known that this information can be trivially re-identified.  Other solutions such as differential privacy require a tradeoff between privacy and accuracy without providing robust cryptographic security.  They need a way to evaluate functions on this customer data at a granular level with arithmetic accuracy, such as specific columns and entries within the database, without seeing the individual data points.

Solution

Secret Computing™ can operate at many levels: across organizations and governments, between departments and jurisdictions within an organization, and at the granular level of segmented customer data within networks and databases. The data science team runs statistical analysis and regression models using XOR, without having direct read permissions on the customer data, thus complying with internal corporate privacy and SOC 2 requirements. The IT Administrator is able to keep all existing security protocols in place, including at-rest and in-transit encryption protocols.

For more detail on the levels at which data can be secretly computed, please reference this infographic.

Family.jpg

Insurance Companies


Insurance Companies

Insurance Companies


Insurance Companies

Fraud detection 

 Coming soon.