Fortanix and Nvidia use Xeureka to improve the accuracy of AI models in drug discovery

A new report released today by Intel Corp.-backed multicloud security company Fortanix Inc. details how Xeureka Inc., a subsidiary of Mitsui & Co. Ltd., used Fortanix's confidential computing technology to improve the accuracy of artificial intelligence models and reduce drug discovery time and costs.

Since its rapid rise in 2022 with the release of OpenAI's ChatGPT, AI has shown great potential in various fields, and drug discovery is one of them. By harnessing the power of AI, pharmaceutical companies can analyze huge data sets quickly and accurately, potentially leading to the discovery of new drugs at a fraction of the traditional cost and time.

Xeureka aims to use computer technology in all aspects of drug development to improve the efficiency and effectiveness of new drug research. The use of AI technologies will be a key focus, as the AI ​​models require large data sets. However, the data used in drug research often contains highly confidential, sensitive and personally identifiable information, and those involved must comply with various global data protection regulations.

To address the problem of data usage, Xeureka has partnered with Fortanix and Nvidia Corp. to demonstrate the ability to integrate disparate data sets from multiple sources while maintaining the security and confidentiality of the original sources. As part of this partnership, Fortanix provides a control plane that enables the management of security policies and orchestrates the use of highly confidential data in a secure environment.

Fortanix's confidential computing capabilities include policy management and orchestration of highly confidential data in a secure environment. They also include data sharing and access control for rapid and collaborative data processing that ensures the privacy of multiple parties without exposing sensitive information.

Modern key management provides consistent encryption regardless of data location in hybrid multi-cloud environments, and workflow management defines, manages and tracks data collaboration process flow to ensure data privacy and security precautions are met.

Nvidia's contribution includes secure enclaves and confidential GPUs for the execution environment. The environment runs data, algorithms, and AI models to ensure that highly sensitive data assets remain secure and private during use.

By partnering with Nvidia and Fortanix to enable confidential data collaboration, Xeureka was able to improve the accuracy of AI models from 65% to 74%, reducing the time and cost of drug discovery. Fortanix also points out that this is a testament to how AI can increase the effectiveness and efficiency of new drug development while improving the lives of millions of people around the world.

“This is a perfect example of the power of confidential computing – rapidly processing extremely large amounts of real patient data required by AI models, while maintaining the highest possible privacy and security,” said Anand Kashyap, co-founder and CEO of Fortanix.

Image: Fortanix