TRANSCRIPT:

My name’s Riddhiman Das. I am CEO and co-founder of Triple Blind, and Triple Blind does privacy as a service. What that means is we enable parties to work together on data that’s sensitive because of regulatory reasons, or competitive reasons, or just plain consumer privacy reasons. And we allow those data sets to be used without violating their privacy. What motivated me to found Triple Blind was the fact that we have so much data that is being stored by enterprises, and in large companies, and banks, and healthcare institutions that just never sees the light of day. The problem Triple Blind solve for companies is one of two, typically. 

One, is either providing access to data that has historically been inaccessible because of the sensitivity around that dataset. On the other side, there are companies that have lots of data that they know is valuable to third parties, but they don’t allow them to access because of the sensitivity around that data. So, this, essentially, Triple Blind allows the latent datasets of those organizations to be monetized safely and securely without exposing them to any regulatory violations or compliance hurdles.

So, how Triple Blind is better than existing solutions is the ability to operate at scale on any kind of data on any operation. What that means is we actually can handle big data. We can do deep neural network training. We can do data processing. We can just do statistical modeling on any kind of data. This can be voice, text, video, tabular data, or images. The reason why it’s faster is because it’s not a CPU-bound isomorphic encryption. It is not hardware-bound like secure enclaves, and you don’t have to use an approximation of the data like synthetic data. This is applicable in medical fields or anywhere where you need to use real data and actually use the outliers in your analysis, as opposed to masking the outliers as you would in synthetic data.

Triple Blind’s mathematical breakthrough eliminates the need for those oblivious transfers. You accumulate less errors as you are progressing. And at the end, we offer the exact same accuracy and precision as if you were operating on unencrypted data. And it happens significantly faster than was ever possible before. In a particular healthcare use case, we enabled our healthcare partner to build an algorithm using EKGs of the heart. In this particular example, we worked with EKGs. We enabled our healthcare partner to build an algorithm that detects atrial fibrillation of the heart 13 years before their best cardiologist’s manual diagnosis. And this algorithm had significant IP and lots of… They spent millions of dollars in building this algorithm, so we were able to encrypt this algorithm using a novel invention by Triple Blind, where this protected the intellectual property and the algorithm as it was licensed to third party countries. And the specific privacy regulation in those countries was also respected because the data that was sent to this algorithm was encrypted and de-identified at the time of sending.

In financial services, what typically happens is that there are data silos. No individual, typically banks, or uses just one financial institution. We may have a credit card at a bank A, we may have a checking account at bank B, we may have an investment account and investment brokers at C. And in those cases, if you just build your anti-money laundering, or anti-terrorism funding, or credit card fraud examples with one of those datasets, you are potentially missing out on the complete financial picture of the person. We’ve been able to protect the privacy of the bank’s customer data as well as the bank’s proprietary information about that customer, while still allowing the data analyst company to have a more complete financial picture of the person, and build a more accurate and a more generalizable credit card fraud detection algorithm.

We’re very excited about Accenture’s investment because it shows that this technology is not just a scientific experiment in a lab. It is a real-world product that can be deployed today. So, Accenture’s investment reinforces to us that our particular technology approach here is credible and is market-ready so that it can be deployed at not just the customers that we sell to directly, but with Accenture’s channel partner here. It really enhances our go-to-market journey. It’s really unusual for Accenture to invest in what is our pre-seed round. We are the youngest investment ever by Accenture in the earliest stages of a company, which really shows that the conviction they had in Triple Blind and its ability to solve this problem.

Some of our angels include Brian McClendon, who was the founder of Google Maps, and Google Earth, and the founder of Uber Maps as well. They include Manik Gupta, the Chief Product Officer of Uber, as well as people that have started and sold technology high-scale businesses in the past. In addition to that, we have had participation from Flyover Capital, a partner that we’re really excited about. The fundamental design principle that we’ve used in Triple Blind is to essentially be the Stripe of privacy. Triple Blind, similarly, is accessible to any engineer or product people with very little overhead. You can get up and rolling in under a day. If you want to try this, and you think Triple Blind can help you access more data, or safely, securely, let others access your data or comply with privacy regulation via technology, please reach out at tripleblind.ai. We’ve got wonderful use cases that you can study to see how it might be applicable in your use case. You can get a demo or you can just drop us a note at contact@tripleblind.ai.