At present, the insurance industry is facing profound transformation. With the promulgation of the "New Country Ten" policies for the insurance industry, more stringent standards and requirements have been put forward to improve the protection capabilities and service levels of the personal insurance industry. This means that the insurance industry needs to bid farewell to the extensive development model and move towards higher quality and more standardized services. During this transformation process, insurance institutions need to continue to improve the quality of insurance services to meet new market and regulatory requirements.
In order to better standardize insurance service behavior, powered by large models, Shuidi launched the "AI Large Model Insurance Quality Inspection Solution", which greatly expanded the coverage of quality inspection, improved work efficiency, and also effectively controlled operating costs.
The traditional quality inspection method is to use manual random inspection, where quality inspectors randomly check the recordings of the entire customer service process and mark out risk content. A quality inspector completes an average of 10-15 inspections of recordings every day, and the quality inspection channels are relatively single and cannot Covering corporate WeChat, public accounts, Moments and other channels is a very labor-intensive task with limited productivity and ceilings.
In recent years, many technology companies have tried to introduce intelligent quality inspection technology into the insurance industry. However, small-model quality inspection is difficult to understand the complex insurance service logic and changes in language expressions, has a relatively low accuracy, and has a high risk of missed inspections.
"Waterdrop AI Quality Inspection" is based on large model contextual semantic understanding and long text reasoning capabilities. It can deeply understand complex conversations, user intentions and emotional attitudes, and identify more hidden and complex quality inspection rules. Internal data shows that Waterdrop AI quality inspection can achieve 100% full coverage, including operational behaviors through voice, corporate WeChat chat records, Moments and other channels. The cost of quality inspection has also dropped significantly. Taking into account the cost of manual review, the cost of AI quality inspection can be reduced by more than 50% compared with manual quality inspection (not taking into account the early development cost).
Behind the "AI quality inspection" capability, it is inseparable from the in-depth empowerment of Shuidi Company's self-developed "Water Drop Water Guard Model". In the past, due to the wide variety of insurance products, relatively complex terms, and lack of massive data for model training, it was difficult to achieve a breakthrough in the accuracy of AI quality inspection.
Shuidi has invested in AI construction since 2019, and has accumulated a large amount of industry segment data under the premise of compliance, including massive insurance vertical service corpus samples, more than 7,000 current or past hot-selling insurance product data, and is proficient in 10,000+ Professional issues in the field of medical insurance. Be able to immediately identify problems such as irregular service attitudes, excessive sales promises, misinterpretation of product coverage, and inadequate insurance purchasing tips. After continuous training, with the manual inspection accuracy rate of 100% as the standard, the accuracy rate of AI quality inspection has been close to the manual level.
According to reports, Shuidi’s “AI large model insurance quality inspection solution” has formed an external export solution and can be opened to the industry. The next step will be to continue to strengthen training on quality inspection accuracy, help more insurance institutions upgrade service efficiency and quality, and promote high-quality development of the industry.