Easy digitization: The application value of privacy computing technology is to realize all-round data circulation on the premise of protecting information security, thus promoting economic development. In recent years, with the initial maturity of technology, the gradual introduction of data regulation and support policies, and the growing market demand for data, the birth of the private computing market is timely and develops rapidly. However, the privacy computing industry is still in the exploratory period, which is embodied in the technological performance to be improved, large-scale application is still being explored, and the industrial development is limited by the quality of data sources and ownership issues.
Analysys divides the development cycle of privacy computing market into four stages, namely, exploration period, market start-up period, high-speed development period and application maturity period. Currently, China's privacy computing market is in exploration period.
Analysis of the development stage of Privacy computing in China is as follows:
Private computing gains Momentum, but Financial users need to Calm down
Exploratory Period (2016-2023)
Privacy computing first appeared in the paper research Category and Development Trend of Privacy Computing published in 2016. That same year, independent private computing businesses began to emerge.
Industry standards, in 2021, mail tunnels successively issued a "privacy calculation (2021), the white paper" and "computation privacy laws and compliance research paper, the criterion for industry to develop systematic privacy compliance to encourage technology application, the next privacy computing applications boundary will expand with ecological technology and data fusion.
In terms of practical application, by the end of 2021, 88 enterprises have successively released more than 100 privacy computing technology products, and the supply of privacy computing industry shows a rising trend. Customers in key areas have successively completed the proof-of-concept and pilot deployment stages, and a large number of commercialization projects have been gradually implemented. The products in the deployment stage account for 48%, mainly focusing on financial, government, medical and communication operations. However, the actual landing cases in 2021 are still few, and the market is still in the supply-driven stage.
From a funding perspective, up to now, 80% of startups in private computing are in the early stages of funding.
In 2022, the case will break out, and the commercial coverage of key areas will be basically completed, and more industries will deploy private computing. On the one hand, data source owners, private computing manufacturers, users of various industries and regulatory authorities will continue to promote the development of private computing technology standards, and on the other hand, they will pay more attention to the compliance supervision of private computing. With the increase of landing cases and the tightening of industry regulation, the market will gradually return to rationality, and there will be more refined classification of privacy computing, and the market will begin to integrate while increasing revenue. In the exploratory period of privacy computing industry, the technology is initially mature and has been applied in the fields of finance, government affairs and medical treatment. However, there are faults between technology and scene, scene and landing, and problems such as data quality and rights confirmation need to be solved.
Market Launch Period (2024-2025)
After the completion of market integration, it is expected that privacy computing technology will become the underlying demand for data processing, helping the full life cycle management of data security, and jointly building a data modeling platform suitable for various application scenarios with blockchain, artificial intelligence and other technologies, gradually moving towards a mature user-driven market. Companies with superior technology and a good ecosystem will dominate the market.
The privacy computing industry, which is in the start-up stage, has made breakthroughs in core technical issues, has been mature applied in many major fields, and its business model has been verified by the market. But there is still room for the combination of technology and application in areas with great potential, such as industry. According to the "14th Five-Year Plan" for big data industry development, "a modern big data industry system with strong innovation, high added value and independent control" will be basically formed by 2025. As the infrastructure of the big data industry, privacy computing will also enter a period of rapid development in 2025. This technology will become a rigid demand of the digital economy, and manufacturers will form a stable and efficient industrial system of the whole life cycle together with upstream and downstream companies.
Rapid Development Period (2025 -)
In 2025, the privacy computing industry products are highly standardized and widely used in many fields. On the one hand, manufacturers will implement rapid deployment of private computing platforms, and on the other hand, they will efficiently meet the personalized needs of users. The private computing market will be driven by demand. However, affected by the development of big data system, there is still a long way to achieve the secure circulation of global data.
Advice for financial industry users
The application of privacy computing in various scenarios of banks has been initially started. State-owned banks, joint-stock banks and urban commercial banks at all levels with rapid development have begun to introduce technical systems such as federal learning and multi-party security to test the water in scenarios such as customer marketing and business risk control. The application of privacy computing in the insurance industry is relatively weak. China Life And Property Insurance has started testing and launching, and the implementation effect remains to be further tracked and analyzed. For securities and other asset management industry, the current case is still lack, still to be developed and landed.
Based on the comprehensive analysis of the current technological development status of privacy computing and its suitability to financial industry scenarios, the financial industry, especially the banking industry, needs to select scenarios with pain points of data circulation business for "calm testing", and spread out from point to line based on feedback to business effects. Specifically, Analysys suggests that financial institutions should give full consideration to the following points when building privacy computing platforms or launching privacy computing business scenarios:
First, private computing technology is overheated. Only by maintaining reasonable expectations and forming a complete privacy computing application planning around business pain points can business value be truly brought into play. Data security chain of whole life cycle management and data application is relatively complex, is the need for end-to-end collaborative solve and create value, privacy calculation in certain degree and range to moderate broken for data security and application, but the data source of compliance is still need to be guaranteed, can form a complete data chain compliance; Data quality still affects the final business effect without compromising model quality with private computing.
Second, in the selection of technical route, different technical route has its own focus and applicable scenarios, according to the scenario to choose the technology application. For example, the cross-business marketing of banks' internal customers adopts privacy settlement and homomorphic encryption methods, while the joint modeling between banks and peers/different industries needs to consider the horizontal and vertical federation applications under different business types. At the same time, after all, different technical routes are also short, so it is necessary to consider the high performance requirements of MPC, and the loss in security of federated learning modeling, the integration of multi-technical routes and the interconnection of heterogeneous platforms should also be considered simultaneously in the application of scenarios.
Third, financial institutions with good technical advantages can try to independently develop a private computing platform suitable for scenarios based on open source projects, but they need to have a comprehensive understanding and tolerance of difficulties such as development cost, deployment difficulty, development cycle and security risk. At the same time, in the selection of open source projects, security protocol support, functional comprehensiveness and learning cost should be considered comprehensively.
Notice: The third party data and other information quoted by Analysys in this article are from public channels. Analysys does not assume any responsibility for this. In any case, this article is for reference only, not as any basis. The copyright of this article belongs to the publisher. Without authorization of Analysys Digital, it is strictly prohibited to reprint, quote or use any content published by Analysys Digital in any way. Any authorized media, website or personal use shall quote and indicate the source of the original text, and the analysis point of view shall be subject to the content officially published by Yiguan digital, without any form of deletion, addition, splicing, deduction, distortion, etc. Analysys digital does not assume any responsibility for any dispute caused by improper use and reserves the right to hold relevant responsible parties accountable.