Secure data analytics startup Opaque Systems Inc. today announced new innovations for its Confidential Computing Platform designed to improve how organizations analyze and interact with confidential data in large language models and artificial intelligence.

The platform enhancements leverage privacy-preserving generative AI and zero-trust “data clean rooms,” which enable organizations to analyze their collective confidential data collaboratively without revealing the raw data. Data clean rooms, or DCRs, are secure environments that allow multiple parties to analyze and derive insights from their combined data without directly sharing or revealing the raw data to each other, preserving the confidentiality and privacy of the data.

With the new release, the Opaque platform also provides safeguards for machine learning and AI models to execute on encrypted data inside trusted executions environments or TEEs, preventing exposure to unauthorized parties.

The Opaque Platform’s DCR capabilities are said to enable secure, multi-party analytics on fully encrypted confidential data secured in TEEs, allowing for queries to be executed on encrypted data without breaching its confidentiality.

For advertisers and marketers, the advancements in Opaque Systems’ Confidential Computing Platform are claimed to provide advantages, such as allowing different advertising agencies or marketing firms to collaborate on sensitive consumer data to measure ad campaign effectiveness, personalized consumer targeting and measurement.

Opaque argues that the uniqueness of its approach is that though the combined data can be analyzed to generate valuable insights, each party can see only the data they directly own. The confidentiality of each party’s data is preserved and valuable cross-organizational insights can be generated without the risk of exposing sensitive raw data to unauthorized access.

“The challenge with traditional DCRs in the cloud is that customers have to trust that their DCR provider, and the other parties analyzing the data, will not inadvertently or purposefully gain access to the raw data being processed,” explained Rishabh Poddar, co-founder and chief executive of Opaque Systems. “With our new DCR offering, organizations can securely collaborate on data within their business ecosystem to tackle a range of important use cases – from detecting financial crime, identifying fraud, to enabling better marketing insights – while ensuring that their unencrypted, raw data is never exposed to anyone.”

Opaque was last in the news in December when it launched a new platform that includes what it says are the latest advancements in confidential artificial intelligence and analytics. The platform was built to unlock use cases in confidential computing to enable data scientists to secure and perform collaborative analytics directly on this encrypted data.

Image: Opaque Systems