Product & technology

Decentriq v3 Preview - Confidential Data Clean Rooms for scale

All industries
No items found.
Key visual with name of article, the Decentriq logo, and the AMD logo
Written by
Nikolas Molyndris
Published on
July 14, 2022

Discover what's coming with our Decentriq v3. The main topic? Data Clean Rooms for scale.

Recommended reading

We are excited to announce the v3 of our Data Clean Rooms in preview release. In this major update of our platform we provide the world’s first production deployment of an analytics platform using AMD SEV-SNP. We pair this new technology with our existing Confidential Computing setup to offer the most scalable data analysis platform using encryption in-use. For the first time,  data scientists can use their existing Machine Learning workloads and frameworks to collaborate on production-size datasets with Confidential Computing guarantees.

During the past few months we gradually deployed the pieces of what we consider to be the future of data collaboration. From a synthetic data node that generates differentially private synthetic data without sharing raw data to anyone, to interactive Data Clean Rooms that allow flexible data analysis workflows with guaranteed data control.

The reception from our users has been exciting for us to see, as more and more of them are moving past the usual security and privacy problems that were blocking their collaborations; to questions about scalability and production-readiness.

With the v3 of our Data Clean Rooms we are moving the cutting edge to the mainstream.

A first for Confidential Computing, powered by AMD

Combining the technologies that we do, and serving workflows and data sizes that our customers expect in production requires us to look ahead in terms of what is possible. So, for the past few months we worked to enable AMD’s brand-new Confidential Computing technology (AMD SEV-SNP) in our platform. AMD SEV-SNP allows us to scale more use-cases to production-level scalability, reliability and resilience. If you are interested in learning more about our technical journey to reach that point, make sure to subscribe to our newsletter as we are releasing a technical blog on it next week.

Unlocking scalable Confidential Machine Learning across industries

As we hinted before, the deployment of this new technology was not something we did in a vacuum. The motivation and the need came from our users as they realized the scale of the collaborations they could enable with our Data Clean Rooms.

Machine Learning on federated patient data

Talking with our clients we hear that the need for data collaboration in healthcare is starting to become imperative. Advances in digital care, as well as in AI and Machine Learning, create immense opportunities for every part of the value chain. From drug discovery, to treatment adherence and clinical trials. Increasingly, our healthcare users are coming up with more and more of these use-cases that depend on advanced Machine Learning capabilities.

Data Clean Rooms v3 allow our users to utilise their usual Python and R workflows and deploy them into production with increased scalability and performance. Data science teams of Life science organisations can combine and analyse data coming from any patient point of contact in the world, without the data owners ever losing control or disclosing patient personal data.

You can learn more about Data Clean Rooms in healthcare by watching our webinar with Roche here.

AI-based first-party data activation without sharing data

As Data Clean Rooms expand in media and advertising, brands and publishers alike are pushing the boundaries of customer insights and audience activation. With v3, Decentriq offers the first Data Clean Rooms that allow marketing teams to combine their first party data with their wider ecosystem, to identify and activate new audiences as well as understand their users better. All while never actually sharing their data.

V3 brings forward the capability to perform even more complex Machine Learning analyses on combined customer data in bigger scale, and faster. For the first time, brands and publishers can build custom lookalike audiences, attribution models, and create (or just copy-paste) custom matching algorithms into a Data Clean Room without ever sharing  customer sensitive data with anyone (including Decentriq).

You can learn more about Data Clean Rooms for Media and Advertising in the latest Forrester report here.

Redefining Trust

Trust is Decentriq’s DNA, and v3 of our Data Clean Rooms is very important for us because it’s the culmination of our commitment to our trust guarantees.

Not only we are the first to bring to production a new Confidential Computing technology from AMD, but with the resulting combination of the available technologies, we massively increase our platform’s scalability and performance. This way we fulfil the promise to provide a safe, neutral space for sensitive data collaboration that caters for the need of the modern data enterprise without any compromises.

The v3 preview release will be made available to selected users in the next few days. If you are also interested in participating in the v3 preview release or you want a demo, reach out to


Recommended reading

Related content

Subscribe to Decentriq

Stay connected with Decentriq. Receive email notifications about industry news and product updates.