How Industrial Bank’s Cloud‑Native AIOps Earned Top‑Tier Assessment

Intelligent Operations (AIOps) applies artificial‑intelligence techniques such as machine learning to IT operations, extracting and analyzing ever‑growing volumes, varieties, and velocities of data to support automated incident detection, diagnosis, and remediation.

At the 20th GOPS Global Operations Conference in Shenzhen on 7 April 2023, the China Academy of Information and Communications Technology (CAICT) announced the first batch of evaluation results for the AIOps General Capability Requirements standard. Industrial and Commercial Bank of China (ICBC) received the highest‑level (Level 3) rating for its "ICBC Cloud‑Native Intelligent Operations Construction Project," becoming the first financial institution to pass the assessment.

The assessment covered the abnormal‑detection and log‑location modules of the AIOps Capability Maturity Model (Part 1: General Capability Requirements). ICBC’s platform now monitors over 150 billion transactions daily, blocks more than 100 k abnormal flows per month, and has automated over 300 k incident handling actions, with more than 70 related patents filed.

Q&A with Liu Chengyan, Senior Expert, ICBC Software Development Center

Q: Please introduce yourself, your organization, and the project evaluated.

A: ICBC is a capital‑strong, internationally recognized commercial bank. In response to rapid digital‑economy growth, we built a cloud‑native intelligent operations system from scratch, establishing three observability pillars—metrics, logs, and traces—and enabling end‑to‑end automated fault detection, localization, and remediation. The platform processes over 150 billion transactions daily, intercepts more than 100 k anomalies monthly, and has handled over 300 k incidents, filing 70+ patents.

Q: How does it feel to pass the AIOps standard assessment?

A: After being the first financial firm selected for the 2022 XOps ecosystem summit, we are thrilled to achieve the highest industry rating, confirming our technical leadership and providing confidence for continued innovation.

Q: What considerations drove your participation in the assessment?

A: We aimed to benchmark our cloud‑native intelligent operations against the standard and to exchange insights with industry experts. The evaluation involved over 14 modules and 120+ items across development, testing, operations, and infrastructure teams.

Q: What changes has the assessment brought to your team?

A: The recognition validates our achievements, boosts confidence, and the systematic review helped us grasp future AIOps trends, guiding the next phase of development.

Q: What are your next steps for AIOps?

A: We will (1) align the AIOps system with ICBC’s four‑fold transformation (modular, unit‑based, domestic, intelligent) to enhance change‑risk control; (2) build a full‑stack root‑cause analysis platform for rapid fault localization; (3) explore large‑model AI to improve ticket handling and intelligent Q&A.

Q: How do you see the future of AIOps?

A: AIOps will shift from an auxiliary role to a core function, driven by self‑learning capabilities that reduce manual data labeling and by the growing power of large AI models, enabling higher accuracy and broader coverage across the operations lifecycle.

For more details, the CAICT contact is Shang Mengchen (wechat: @尚梦宸, phone: 132‑6108‑1232, email: [email protected]). Additional contacts include Wei Huanxin from the GreatOps community.