Trusted-Environment-based Cryptographic Computing (TECC): Patent Authorization and Performance Advances

Trusted-Environment-based Cryptographic Computing (TECC) is an innovative privacy‑computing technology originated by Ant Group. Recently, TECC received a patent authorization for “Distributed Multi‑Party Secure Computing System, Method and Node” (Patent No. CN113992439B).

In the era of data in encrypted state, privacy computing is considered a key technology that simultaneously enables data value circulation and data security protection, but current mainstream privacy‑computing techniques—whether multi‑party secure computation, federated learning, trusted execution environments (TEE), or fully homomorphic encryption—have security or performance limitations, especially the trade‑off between computational efficiency and data protection, which is a major obstacle to large‑scale industrial adoption.

Since 2021, Ant Group has been developing the next‑generation privacy‑computing technology TECC, innovatively integrating cryptographic techniques (MPC, FL) with full‑stack trusted computing (TEE, TPM), overcoming the limitations of single‑technology solutions and achieving higher overall capability.

The core breakthrough of TECC is the use of high‑speed fully encrypted computation in remotely verified TPM/TEE environments. This improves performance, reliability, and applicability compared with traditional cross‑network privacy computing (MPC/federated learning), while effectively resisting supply‑chain attacks, side‑channel attacks, and plaintext data leakage in TPM/TEE, as well as collusion, malicious adversary, and information‑entropy leakage attacks in MPC and federated learning.

Compared with other privacy‑computing technologies, TECC is more suitable for scenarios with high data‑security requirements (e.g., critical data), large data scale (e.g., millions of personal records), complex computation logic, and variable numbers of participants, and it also supports cross‑regional data‑center encrypted computation.

During its development, TECC has produced multiple advantageous technologies and obtained six patent authorizations within a year. The recently authorized “Distributed Multi‑Party Secure Computing System, Method and Node” can increase TECC's computation speed by 10‑ to 100‑fold depending on resource consumption and task parallelism, achieving billion‑scale encrypted GBDT model training within one hour and billion‑scale encrypted SQL analysis within ten minutes, delivering a computation experience comparable to plaintext processing and representing a significant breakthrough in privacy‑computing performance.

“Distributed Multi‑Party Secure Computing System, Method and Node” technical route

The comprehensive advantages of TECC can meet industries with massive data volumes such as energy and industrial internet, as well as large‑scale computing scenarios like “East Data West Computing”, providing a forward‑looking technical layout for broader privacy‑computing applications and is the most promising technology to satisfy the demands of the encrypted‑state era.

Ant Group has completed the theoretical framework of TECC and applied it in privacy machine‑learning training, offline batch prediction, real‑time prediction, and encrypted data analysis.

Meanwhile, TECC is undergoing professional evaluation by institutions such as the China Academy of Information and Communications Technology (CAICT) and BCTC, and will gradually be open‑sourced under the guidance of regulatory authorities, inviting industry participants to jointly promote ecosystem construction.