Ant Group Presents AI‑Driven Threat Detection Using Parallel Security Slices at the 2nd Wuhan Cybersecurity Innovation Forum

On April 23, the second Wuhan Cybersecurity Innovation Forum, hosted by the Wuhan Municipal Government and supported by the China Internet Development Foundation, opened and announced the "2024 Top Ten Outstanding Cybersecurity Innovation Achievements," with Ant Group’s "Aspect‑Fusion Intelligence in Threat Detection" selected.

During the innovation sharing session, Ant Group senior algorithm expert Zhong Zhenyu explained the solution, emphasizing the growing complexity of cyber threats in the era of digital transformation and the need for real‑time, precise detection and rapid response.

Ant Group innovatively combined artificial intelligence with its proprietary security parallel‑aspect technology, applying a large‑model security framework (DKCF – Data/Knowledge/Collaboration/Feedback) to create a multi‑layered, depth‑defense system centered on "parallel security aspects and intelligent threat detection."

The system includes four key modules: (1) Data correlation analysis, which builds complete attack chains using aspect‑based data; (2) Unknown threat discovery, leveraging the DKCF framework to apply the large model’s knowledge abstraction and reasoning capabilities; (3) Alarm false‑positive reduction, matching alarm data with business behavior baselines to cut invalid alerts; and (4) Security‑countermeasure knowledge graph, integrating ATT&CK tactics, threat intelligence, and knowledge‑graph nodes to enhance detection accuracy and explainability.

This architecture addresses common shortcomings of existing threat‑detection methods—high false‑positive rates, weak unknown‑threat discovery, and poor interpretability—demonstrating superior performance in drills and real‑world practice, and earning recognition such as the 2024 WIC Find Intelligent Technology Innovation Case, inclusion in the "2024 AI Pioneer Cases" collection, and selection as a model project for large‑model application.