Current location - Trademark Inquiry Complete Network - Tian Tian Fund - Cooperate with PICC P&C Insurance Company on fixed losses.
Cooperate with PICC P&C Insurance Company on fixed losses.
It took the insurance company 10 years to realize the electronization of only accident photos and claims data from the initial camera survey to the later application of claims system and now the mobile survey. But this year, great changes have taken place.

Ant Financial, China Ping An, China People's Insurance Company and China Baoxin Company cut into the field of fixed loss from different angles with science and technology.

Among them, Ant Financial and China Ping An focus on the efficiency of loss assessment, and the new intelligent loss assessment tool aims to greatly improve the efficiency of simple case claims. When the fixed loss efficiency of a simple case reaches the second level, there is really no need to wait for the repair of a simple case in the future.

People's Insurance Company of China has developed and purchased a variety of accident vehicle detection tools, trying to find vehicle losses without disassembly inspection, and paying attention to the quality of claims. China Baoxin relies on the most comprehensive database in the industry, focusing on anti-fraud, which is expected to reduce the probability of fraudulent insurance.

Now, this intelligent loss-fixing competition has begun. Who will have the last laugh?

The contest between fixed loss treasure and AI intelligent flash compensation

Ant Financial is the first giant to announce the empowerment of artificial intelligence this year. In June, the company launched a fixed loss treasure, which gave it another identity-a technology provider in the insurance industry.

According to the interface news, Dingbaobao uses artificial intelligence to simulate the manual operation process of auto insurance loss determination to help insurance companies realize automatic loss determination in simple cases. Behind it is the application of picture recognition technology in the insurance field. As the eyes and brain of the loss adjuster, AI identifies the accident photos through algorithms to determine the damaged parts and the degree of loss. Then, by docking with the maintenance price library of the cooperative insurance company, the damaged parts, maintenance plan and maintenance price can be given in a few seconds.

This is the first time in the field of auto insurance to realize the commercialization of image loss determination technology. However, this loss assessment tool, which is known as "claim 65,438+10,000 lives", has been questioned by many people in the field of claims after its launch.

The first problem is the massive historical pictures. In order to achieve the effect of image recognition and damage determination, it is necessary to learn a large number of historical auto insurance accident photos by machine. "70% of the premium income of the auto insurance industry is in the hands of China PICC, China Ping An and China Pacific Insurance, which means that about 70% of the insurance claims data are in the hands of these big companies." Weasion Wei, the founder of A6 Studio, a senior claimant, said that these are the core data of insurance companies, and all three companies are capable of data analysis, so it is unlikely that * * * will enjoy Ant Financial Services. In fact, China PICC, China Ping An and China Pacific Insurance are all promoting artificial intelligence.

Insiders of Ant Financial Insurance Group told the interface news that the training machine really needs a lot of historical photos of auto insurance. Large and medium-sized insurance companies such as Dida Property Insurance and Taiping Property Insurance have cooperated with machine learning and have enough historical data for machine learning. The person further added, "Different insurance companies have different models and photo shooting specifications. Some companies shoot more flat shots, some prone shots, some emphasize detail shooting, and some like to make annotations. We use the photos of various insurance companies to' feed' machine learning, so that the fixed loss treasure covers all models and adapts to photos in different weather and different shooting methods. "

Another question that has been questioned more is the localization of parts prices. The most important thing to determine the damage is to determine the maintenance price, which involves the prices of accessories, accessories and working hours in various regions of the country. As an Internet finance company, Ant Financial obviously does not have this data. In this regard, the insiders of the above-mentioned Ant Financial Insurance Group said, "In essence, the fixed loss treasure does not involve this part. Fixed loss treasure only does the picture identification and fixed loss in the previous paragraph, and the system will automatically submit the results to the insurance company. The pricing power is still in the hands of insurance companies, and each insurance company determines the price according to its own price library and outputs the maintenance plan. "

In addition to technical doubts, the industry is most worried about moral hazard. Some senior claim adjusters interviewed by Interface News generally said that they did not see the anti-fraud function of Dingbaobao, and worried that fraudsters would take advantage of this loophole and bring risks to the risk control of insurance companies.

According to the interface journalist, after discussing with the cooperative insurance company, Dingbaobao has some anti-fraud designs on the logical relationship of accessories and the authenticity of accidents, which are used to judge the authenticity of accidents and warn some cases that do not conform to the logic of missing accessories. However, the above-mentioned Ant Financial Insurance Group also admitted that only some simple anti-fraud can be done at present. "It will continue to deepen in this direction in the future, which is a process of continuous optimization," he said.

The days when Ant Financial monopolized the field of intelligent loss determination in the voice of doubt did not last long. Last month, Ping An of China announced the launch of its own intelligent loss assessment product-AI intelligent flash compensation, which is the first intelligent loss assessment tool launched by insurance institutions.

Similar to the loss-fixing treasure, AI intelligent flash compensation also adopts image recognition technology, and the machine learns massive historical real claim image data training to realize image recognition and loss-fixing. Interface news learned that "it has a lot of historical car damage picture data" is the main advantage of this part. China Ping An owns 19.83% of the property insurance market, which means it has about 20% of the industry's car damage data. China Ping An publicly stated that AI smart flash compensation is the only artificial intelligence risk control product put into use in real production environment.

However, after careful comparison of interface news, it is found that the biggest difference between fixed loss treasure and AI smart flash compensation lies in the automatic pricing link. The pricing work of the fixed loss treasure is transmitted back to the insurance company, while the AI smart flash compensation is based on Ping An Property Insurance, and the working hours and parts price system covering the whole country and accurate to counties and cities is built, and the price is set by itself. For this reason, AI smart flash compensation is the first intelligent tool to integrate fixed loss pricing.

In addition, AI smart flash compensation has not been questioned by "moral hazard". The reason is that when AI smart flash compensation is released, it is revealed that it has built a total risk factor library from underwriting to claim settlement, covering the main "case" and "help" risks in the whole process of claim settlement, and realizing intelligent interception and intelligent screening of post-event risks.

The person in charge of Cailletet Ming Technology R&D Center, a third-party risk control company cooperating with Ping An, told Interface News that they hired an experienced mechanic to write down the experience of vehicle maintenance, and the experience of accidents every day became a new rule. "People's experience is written down and compiled into computer logic rules. These logic rules become interception rules when they are put into the system. There are currently more than 40,000 rules. "

"These rules are like spiders weaving webs. Under normal circumstances, it is leaked from the power grid. When a problem case triggers a rule, it will be intercepted and the adjuster needs to take it out and look at it carefully. " The person in charge said that this set of rules is the most effective for professional insurance fraud interception in auto repair shops, because they cheat insurance according to the logic of vehicle maintenance and body structure.

These intelligent damage assessment tools consume a lot of manpower and material resources on both sides. China Ping An disclosed that in the past five years, the annual innovation investment in the field of financial technology was nearly $654.38 billion. Industry insiders estimate that the investment of Ant Financial should also be in the level of 100 million yuan.

Interestingly, Ant Financial and China Ping An both gave their fixed loss products to insurance partners for free. According to Ant Financial, there is no charge for this project at present, and China Ping An will give the cooperative organization a three-month free period. It is predicted that the industry may move towards the charging mode in the future, and technology output will become a new profit point for China Ping An.

But there is no free lunch in the world. The purpose of these two free fixed-loss products is to run data. "As far as I know, Ant Financial and China Ping An's fixed loss product solutions are the same logic." Weasion Wei told Interface News that the two companies gave the tools to other insurance companies for free, in order to run more insurance data and make the damage assessment results of the tools more and more accurate.

According to the interface news interview, between Ant Financial and China Ping An, insurance companies are more willing to cooperate with Ant Financial. At present, dozens of insurance companies have expressed their intention to cooperate, and Ant Financial is arranging for docking one after another. China Ping An previously announced that more than 10 insurance companies have signed a letter of intent for AI smart flash compensation cooperation.

A person in charge of the auto insurance business of a small insurance company explained to the interface news, "China Ping An is a competitor of the insurance company after all, and it is somewhat worrying to hand over the company's core claims data to the competitors. Ant Financial is positioned as a technology company. Insurance is only a part of its entire ecology, and it is more of a gesture of serving insurance companies, which will make insurance companies more assured. "

Another advantage of Ant Financial is that it has a lot of "human data", which insurance companies lack. For example, Ant Financial launched the "Auto Insurance Score" in May, which can make accurate portraits and risk analysis of car owners. It is understood that there are 18 cooperative insurance companies with auto insurance protection in the first 100 days, and the average daily call frequency is 1 10,000. Auto insurance points are mainly used for auto insurance pricing. AXA Tian Ping, the first company to launch auto insurance points, disclosed that after the launch of auto insurance points, 65% of users' risk identification results have changed.

Since the fixed loss products of the two companies went online, there have been constant discussions about PK of the two companies. Yin Ming, president of Ant Financial Insurance Group, told Interface News: "Ant is a platform and will not compete with insurance companies. On the contrary, we welcome insurance companies to join the insurance technology field. "

In Yin Ming's view, the ultimate winner of insurance technology is not an institution, but the whole industry and insurance consumers.

This is not a competition.

Unlike Ant Financial and China Ping An, which cut into the field of fixed loss from intelligent software, PICC China deeply cultivated fixed loss hardware.

This leading property insurance company launched six scientific and technological claims tools in September, including two customized products of PICC P&C, one jointly developed by PICC P&C and manufacturers, one product with national invention and innovation patents invented by PICC P&C claims experts, and two outsourced products of PICC P&C. ..

Among them, the most important is the engine connecting rod disassembly-free measuring tool, which rewrites that wading vehicles must disassemble the engine to judge the wear history. Independent research and development by PICC auto insurance claims department, with national utility model invention patent. If the owner of the flooded vehicle uses this disassembly-free measuring tool, the engine can be prevented from being disassembled during vehicle maintenance, and whether the engine connecting rod is damaged can be accurately detected within half an hour to five hours, which not only reduces the waiting time for maintenance, but also maintains the original state of the engine.

The developer of this tool told the interface news that it is a complicated case that the wading accident vehicle involves engine loss. The traditional detection method must disassemble the engine, which wastes a lot of manpower and material resources and consumes a lot of time to determine the damage. "More importantly, in the past, the right to speak was in the hands of repairers, and neither insurance companies nor car owners could stop them from dismantling their cars for their own benefit. With this tool, car owners and insurance companies have the initiative. "

In addition, PICC P&C has customized two exclusive devices-electronic control system diagnosis device and cloud diagnosis device. The electronic control system diagnostic instrument is used to measure the daily loss of the electronic control system, and the fault range can be quickly determined, the fault type can be judged and the fault reason can be analyzed through the on-board diagnostic port. Cloud diagnostic equipment is mainly aimed at high-end vehicles, compatible with many imported and joint venture high-end brand models, and can realize remote control and detection. For example, if the car engine is out of the cylinder or the light is not on, you can plug these two devices into the on-board diagnostic port, scan them and get the results in a few minutes. Especially for high-end vehicles, there is no need to wait in line in the garage when testing.

"These tools independently developed and purchased by PICC P&C have broken through the technical barriers of the auto repair industry chain, and also innovatively mastered the technology of' no disassembly and no inspection', which can quickly handle large and complex cases, especially high-end car cases in the future, which may form the core advantage of PICC P&C in fixed loss claims." An industry insider analyzed.

People from PICC P&C Claims Department told Interface News that although the damage determination is very complicated, there are still laws to follow. This series of tools is to standardize the loss determination behavior. In the past, manual loss determination mainly relied on the experience of each adjuster, which was difficult to integrate. "Now through the way of image loss determination and intelligent loss determination, it has become information that can be collected and sorted out, and it can be analyzed and managed in the background. This is the biggest role."

"Through data analysis, we learned about the parts of each model that are prone to accidents, and also found the parts that need cost control. This is not aimed at customers, but affects dangerous parts and even coordinates with the maintenance industry. The background of an insurance company controls the cost of a certain type of accident and a certain type of vehicle more accurately, which is the most costly management. " He added.

Compared with Ant Financial, China Ping An and China PICC, China Insurance Information Technology Management Co., Ltd. (hereinafter referred to as China Insurance) is more like an industry platform.

China Baoxin, established with the approval of the State Council in early 2004, is known as "UnionPay" in the insurance industry. Funded by China Insurance Guarantee Fund, it was born to strengthen the infrastructure construction of the insurance industry and comprehensively improve the information level of insurance management.

Auto insurance data is the first data integrated by China Baoxin. Because of this, China Baoxin holds the most comprehensive claims data in the industry. With the advantage of data, China Baoxin launched "Fixed Loss Cloud".

Fixed loss cloud includes accident car fixed loss system, anti-fraud system, statistical analysis system and other application systems. After the insurance company cooperates with China Baoxin to use the fixed loss cloud, it can find the data of vehicle fixed loss, parts selection and nuclear loss pricing. Different from the powerful functions of fixed loss products of other commercial giants, the positioning of fixed loss cloud is an auxiliary operation platform and ecological platform for auto insurance claims.

According to the news interview on the interface, the industry generally regards the fixed loss cloud as a supplementary tool for the fixed loss link, and attaches great importance to its application in anti-fraud. In the past, the insurance data standards and data formats used in different places were not uniform. After the establishment of China Baoxin, it broke the data barriers of regions and insurance companies, standardized the collection of 54 anti-fraud data elements and unified the data standards.

Based on the application of auto insurance platform data, China Baoxin developed an auto insurance anti-fraud information system, built its own risk database, and realized the functions of insurance fraud clue identification, fraud risk early warning and fraud information sharing, which provided information and data support for companies to crack down on insurance fraud, especially for cross-company fraud cases that are difficult for a single insurance company to identify.

Like Ant Financial and China Ping An, China Baoxin, as an industry platform, is also open to companies in the whole industry. Previously, the fixed loss cloud has started to test the data of the first batch of 10 insurance companies. The person in charge of the auto insurance business of the above-mentioned small insurance companies believes that as the industry platform with the strongest data, China Baoxin will attract more insurance companies.

The entry of the Big Four has benefited the whole insurance industry. The experiment of picture recognition in the field of auto insurance has begun, and its application in other types of insurance in the future is also worth looking forward to. For insurance consumers, no matter what kind of service they enjoy, it can bring about an improvement in the claims experience.

According to Zhong Cheng, president of Anxin Insurance, several companies have their own advantages. China Ping An's AI intelligent flash compensation connects image recognition and background database well, which solves the uncertainty and standardization problems in loss determination, reduces online and offline claims, and improves efficiency. At the same time, the intelligent interception function of AI intelligent flash compensation will intercept the problem case. At this time, China Baoxin's industry data system will be used to query the historical accident data of vehicles, supplemented by the owner's auto insurance credit, so as to make a more reasonable judgment. The most important thing is that insurance companies should use these tools in a down-to-earth manner in order to see their role in risk identification and process.

"Insurance companies should use it, and we plan to cooperate with Ant Financial, China Ping An and China Baoxin," Zhong Cheng said.

Further reading: How to buy insurance, which is good, and teach you how to avoid these "pits" of insurance.