(Hong Kong, Shanghai, 16 July 2020) Ping An Insurance (Group) Company of China, Ltd. (“Ping An” or the “Group”, HKEx:2318; SSE:601318) is pleased to announce that the Ola-Bond Default Analysis developed by Ping An Technology (Shenzhen) Co., Ltd. (Ping An Technology) won first place in the financial direction scenario challenge of the Artificial Intelligence World Innovations Challenge at the recent World Artificial Intelligence Conference (WAIC).

The Ola-Bond Default Analysis uses AI to assist bond investors conducting risk management before, during and after investment. The five O-RISQ models developed by the Ola-Bond Default Analysis achieved an accuracy rate of 93%.

• The Operation Oriented Model provides users with customized and thematic analysis modules. Users can conduct in-depth research on specific topics or multiple types of topics independently.
• The Risk Diffusion Model identifies the risks associated with the bonds. It builds a relationship diffusion network map of the associated bonds for bond issuers, underwriters and guarantors. This helps to calculate the probability of incoming risks.
• The Intelligent Expert Model is based on massive research reports and experts’ in-depth analysis methods to refine expert experience, generate an expert knowledge base and send bond risk reports to users through simulated expert analysis methods. 
• The Similarity Evaluation Model calculates the degree of similarity regarding the risks of the underlying bonds based on the historical default cases database. It constructs a comprehensive database of default cases, matches the features of default bonds with the underlying bonds and flags cases of similar default risk.
• The Quantitative Risk Model provides multi-dimensional risk scanning results and generates a rating based on the factors, incidents and dimensions of the bonds, which enables users to quickly identify the reasons for a bond default.

The Ola-Bond Default Analysis has four characteristics:

• Wide coverage: comprehensive coverage of credit bonds, approximately 220 million+ social entities, 1 billion+ people, entities, industries, incidents and relationship nodes.
• Speedy updates: rapid extraction of risk clues in less than 3 seconds, and data updated once per hour.
• Smart monitoring: 24/7 in-depth coverage of the entire process of investment risk control monitoring, which accurately sends risk clues.
• Fully associative: construction of a full range of associated paths based on the bond subject, solve the problem of macro-medium-micro analysis of fault, and fully explore the risk of bond default.

Since the Ola-Bond Default Analysis was established in 2019, it has provided users with early warning of potential defaults of bonds worth more than USD21.4 billion (RMB150 billion). Compared with traditional early warning systems, the Ola-Bond Default Analysis provides warnings four months in advance, which enables each aspect of bond risk control to be more efficient.

Dr. Xiao Jing, Chief Scientist of Ping An Group, said, “Thanks to the core technologies of AI and our ‘finance + technology’ strategy Ping An continuously strengthens its foundation and aids the development of the industry. While focusing on how technology boosts our efficiency and capacity, we also pay close attention to risk management and control, in order to protect investors through the power of technology.”

The WAIC conference was jointly organized by the National Development and Reform Commission, the Ministry of Science and Technology, the Ministry of Industry and Information Technology, the Cyberspace Administration of China, the Chinese Academy of Sciences, the Chinese Academy of Engineering and the Shanghai Municipal People’s Government. As an official competition of WAIC, the Artificial Intelligence World Innovation Challenge has held competitions in the field of medical, autonomous vehicle, robots, finance and manufacturing since 2018. Over 1,000 teams from around the world register for the competition each year, including businesses and institutions such as Alibaba, Tencent, Microsoft, Shanghai Jiao Tong University and Shanghai Fudan University.