Cape Privacy applies ML to encrypted data to address security concerns

Join online with today’s leading executives at the Data Summit on March 9th. Register here.

Look back to the days when HTTP / SSL was the newly introduced protocol.

Initially, it was implemented by organizations that shuttle the most sensitive data back and forth – usually credit card or other financial information.

But today, it is in universal use, applied almost everywhere. In fact, if you don’t see that little lock icon on the left side of your address bar, it’s a red flag to exit.

Cap Privacy sees secure multilateral calculations taking a similar adoption turn so that it becomes “absolutely ubiquitous”.

Cape Privacy CTO and co-founder Gavin Uhma said, “We believe in a world where intelligent products and services using AI will be encrypted by default.” “To get there, we need to reduce the cost of adopting secure calculations, making them easy to use and accessible.”

According to IDC, 68% of data collected by organizations goes to waste. One of the primary reasons for this is data security concerns.

Making predictions based on encrypted data

“This is a huge missed opportunity for businesses, and that’s where Cape Privacy wants the industry to be at the forefront,” Uhma said. Its platform enables organizations to use encrypted data on encrypted data without ever decrypting and to run predictive machine learning (ML) models. The San Francisco company has just released a self-service version of the platform that is optimized for the Snowflake Data Cloud.

“It’s important to prioritize security when dealing with the world’s most sensitive information, such as financial data,” said Che Vijesingh, CEO of Cape Privacy. “The potential benefit to the customer when using that data for insights cannot be overlooked. By giving organizations the ability to maintain encryption at all times, they can close that risk gap and finally put their most sensitive data to work safely,” he said.

Traditionally, to collect value from encrypted data, it had to be decrypted, placing plain text data at risk of dangerous actors or human or technical errors.

With cap privacy, data is never decrypted, so there is no risk of breach, Vijesingh explained. The company’s technology – which it called the first of its kind in self-service capabilities – combines a well-established cryptography scheme of secret sharing with secure multi-party calculations. This allows organizations to secure encryption keys and keep sensitive data encrypted even when operating through their chosen prediction model from their snowflake environment.

“It boils down to the ability for users to move sensitive data to the cloud,” Vijesinghe said.

Partnership with Snowflake

Focusing on Snowflake allows for a “focused surgical application”, and the partnership is due to the fact that the data cloud company wants to expand into financial services. The industry is a fast growing adopter of ML and is actively investing in cloud migration.

Up to this point, large financial companies and capital markets have been hampered by federal regulation. But platforms and technologies like Cap Privacy allow them to develop securely and use the data they collect but could not otherwise use it, Vijesinghe said.

Cape Privacy has also partnered with Tablo and DataRobot.

“We know we’ve hit a really strong, focused use case where they need access to all this confidential data,” he said. “Cap privacy allows them to use that data while maintaining security and privacy.”

Vijesingh called it the next level of encrypted data access. For a long time, companies have had to resort to one of two completely different options: moving data to the cloud, then locking it tightly, or using “inadequate technology” such as basic tokenization to make predictions.

“Privacy and security are the top concerns of all the organizations we work with,” Vijesinghe said. Now they can run powerful AI predictions on encrypted data at the point of encryption, he explained. Wherever you apply ML, you can apply secure ML.

The Cap privacy platform is simple when it comes to both setup and use, he said. “It’s a very intuitive user interface with plug and play aspects,” Vijesingh explained. Users can quickly and easily access encrypted data, use it, run queries, create new setups, and add and invite other users.

One of the ultimate goals is to enable organizations to securely and securely share encrypted data with third parties. But securing the ubiquity of multi-party computing is not going to be the “success of the night,” he insisted. Challenges such as complexity, use case application, and scale will unfold over months and years.

“Data sharing is really downstream or upstream where we are today,” Vijesinghe said. “The huge opportunity we see on the horizon is secure data sharing and collaboration.”

Cap privacy goals are no less than high. “Imagine a world where all the data in the cloud is encrypted in some shape or form,” said Vijesinghe.
The company aims to be ubiquitous in that area, which is the way to go, he said. “We believe that, based on the technology we have created, once we scale up and scale out, we become that effective secure engine for encrypted data.”

Venturebeat’s mission Digital Town Square is set to become a place for technical decision makers to gain knowledge about the changing enterprise technology and practices. Learn more

Similar Posts

Leave a Reply

Your email address will not be published.