Enabling trust where none exists, securing collaboration for a safer SOCIETY
RETRIEVE DATA FROM PEERS WITHOUT REVEALING WHICH INFORMATION WAS QUERIED
TRAIN AND QUERY models on highly sensitive data and documents
INFER DATA-DRIVEN INSIGHTS without sharing data with PEERS
ensuring privacy, trust, and global problem-solving at scale, WE enable secure and seamless collaboration in distrusted environments by leveraging Multi-Party ComputATION and the nearly infinite compute power of Sky Computing.
A world where trust is no longer a barrier to solving existential threats, where organizations can securely unite to protect humanity and build a safer future.
WE ENVISION a secure digital landscape for HUMANITY
enable secure multi-agency counterterrorism collaboration by allowing international intelligence agencies to jointly analyze sensitive data without exposing raw information, maintaining comprehensive visibility into global terrorist networks.
securely store sensitive information in compliance with regulations while maintaining accessibility for authorized use, employing encryption and access controls that ensure data remains protected throughout its lifecycle yet retrievable when legally required.
identify criminal connections while protecting innocent individuals' privacy and maintaining legal compliance with data protection regulations.
enable chatbots to securely process and provide insights from RESTRICTED information across multiple organizations without exposing sensitive data, preserving confidentiality while allowing AI-powered analysis of protected intelligence.
OpenMPCC (Open Multi-Party Confidential Computing) is an open-source project dedicated to building a platform for secure, collaborative computation. We combine cryptographic Multi-Party Computation techniques with the scalability and flexibility of cloud computing, leveraging hardware-based Confidential Computing technologies. Our goal is to enable joint computations on sensitive data, even when you don't fully trust the underlying hardware, software, or your collaborators.
In today's data-driven world, collaboration on sensitive data is crucial. However, traditional methods often require sharing raw data, which poses significant privacy and security risks. OpenMPCC addresses this by allowing computation on encrypted data, ensuring that
- Data remains confidential: No party, including the hardware vendor or cloud provider, can access the raw data.
- Collaborators only see the results: Input data from other parties is protected.
- Trust is minimized: We design the platform to function even in environments where hardware and software infrastructure, as well as collaborators, are not fully trusted.
WE AIM for the holly grail in secure computation. We minimize the trust assumptions to the bare minimum overcoming TODAY'S TECHNOLOGY barriers REQUIRING to trust hardwaRe vendors, operating systems, and CLOUD Providers per SE.
OpenMPCC leverages a combination of technologies, including:
- Multi-Party Computation: Cryptographic protocols enable multiple Nodes to jointly compute a function over their private inputs without revealing those inputs.
- Confidential Computing: We integrate hardware-based security features to further protect data and computations from unauthorized access in the cloud.
- SKY Computing: We utilize the elastic compute resources of cloud platforms to scale MPC computations efficiently.
OpenMPCC aims to support a wide range of MPC protocols. The specific protocols supported will evolve as the project develops. Check the project documentation for the most up-to-date information.
We aim to support various TEE technologies, including Intel TDX, AMD SEV, and NVIDIA CCA. The level of support for each technology will depend on community contributions and project development.
OPENMPCC iS AN OPEN SOURCE PROJECT. We welcome contributions from developers, researchers, and anyone interested in secure computation. You can contribute by Developing new features and protocols, Improving documentation, Reporting bugs and issues, OR Providing feedback and suggestions AS WELL AS Spreading the word about OpenMPCC.