- Trustless access control to encrypted content (winner ofhttps://www.zama.ai/post/zama-bounty-program-season-6 )
- Gathering and certification of data from multiple devices (technically: decentralized PKI and validation)
- Trustless validation and auditability of data transfers between distrustful devices, with resilience to corruptions or crashes
- Remote encryption of data at rest in an untrusted cloud (AWS),remote decryption (only when needed) to process it remotely. Secured with sovereign keys (with fault-tolerant cryptographic techniques and hardware security modules).

- Auctions with hidden offers for splitting of a limited resource between distrustful participants (securitytoken, NFT/Initial Public Offering, 5G bandwidth, timeslots for rail or airports).
- Trustless platform for automatic sharing of the resource
- Financial stress tests and statistics across European institutions
- Statistics (financial, accounting) across international subsidiaries of your group without requiring them to share their data
- price making without disclosing individual preferences
- KYC, compliance and fraud detection: collaborate with your concurrents to increase your detection accuracy, without showing the data of your customers
- Secure outsourced computations in the cloud: externalize your applications and your data in the cloud but do not let anyone see it
- Keep ownership of the encryption keys and process your data privately in the cloud where it is stored
- Tax optimization without asking your client to show their financial data in clear

- Prevention of sick leave: predict transmission of diseases in your company and risks of long term sick leave, in collaboration with your insurer, while guaranteeing privacy of your employees data
- Secure outsourced computations in the cloud: externalize your applications and your data in the cloud but do not let anyone see it. Keep ownership of the encryption keys and process your data privately in the cloud where it is stored
- Train a collaborative model with your partners without seeing their data (private federated learning, private aggregation of models, full custom private training)
- Secure usage of the model: the model is encrypted so predictions can only be done if all partners are ok, case by case

- Take action based on detailed information of your subcontractors (financial, logistic) without requiring them to reveal their data

- Use medical / DNA information from several countries and traina collaborative model, while keeping the data confidential
- Secure medical messaging encrypted end-to-end between doctor and patient, powered by cheap HSM and fragmented sovereign keys
- Remote encryption of collaborative health database in an untrusted cloud (AWS), then remote decryption in an ephemeral elastic VM with access control on who can process it (but not download it)

- Mutual army recognition (radio signal, images of vehicles)
- Collaborative intrusion detection without showing the logs to each other
- Collaborative phishing/fraud detection without opening the emails / sms / calls history