Our reporter Meng Ke
Trainee reporter Yang Jie
Users of the same Internet platform, VIP users with higher loyalty and stronger spending power pay much more for a service than new customers, and even the service quality is not as good as new customers.
I believe many people have encountered such a situation of "different people have different prices".
Behind this is actually the unreasonable application of the algorithm to "trouble".
In recent years, while the application of algorithms has injected new kinetic energy into social and economic development, problems caused by the unreasonable application of algorithms such as algorithm discrimination, "big data killing", and induced addiction have also profoundly affected the normal communication order, market order and society. order.
Against this background, four departments including the Cyberspace Administration of China recently issued the "Regulations on the Administration of Algorithm Recommendations for Internet Information Services" (hereinafter referred to as the "Administrative Regulations") to regulate "big data".
The industry believes that the promulgation of the "Administrative Regulations" means that the industry related to algorithm recommendation has begun to enter the era of strict supervision.
Prices can be "customized"
The more familiar the customer, the higher the price
"Securities Daily" reporters entered "big data" on a consumer complaint service platform, and jumped out 2073 complaints, including mainstream e-commerce platforms, takeaways and online car-hailing platforms.
A consumer who complained on the platform said that when using an app to buy a meal package, the same time and the same package, different mobile phones will display different amounts; there are also consumer complaints that the price is listed on the page when booking a hotel online. It was 87 yuan per night, but the price dropped by 5 yuan after the payment was complete.
In addition to ordinary consumers, there are also VIP members of some Internet platforms who "spit out" and do not enjoy better treatment.
A consumer said that when buying a router on a shopping software, the purchase price displayed as a member account is 649 yuan, but it only costs 579 yuan for an ordinary account.
In addition, the reporter combed the consumer complaint records and noticed that many large Internet service platforms have user feedback "big data".
An industry insider engaged in big data-related work revealed to a reporter from Securities Daily that the current phenomenon of "big data killing" does exist in large Internet companies. In addition to using their own accumulated data, they also purchase and collect data externally.
In order to find out, the reporter of "Securities Daily" conducted some experience.
On a social platform, the account of user Mengmeng (pseudonym) is already a VIP member, and the reporter himself has applied for a new account. In the same case of purchasing a one-month membership, the reporter only needs to spend 12 yuan, and the platform will give an additional gift For a one-month membership, Mengmeng needs to spend 15 yuan, and the platform only gives away 15-day membership.
Obviously, the two were "differently treated" by the platform, and Mengmeng was "killed" by the platform.
Wu Qi, executive director of Wuxi Digital Economy Research Institute, told the Securities Daily reporter that the reason why "big data kills familiarity" is repeated, from the perspective of platform merchants, regular customers are basically members, who can spend money or pass Most of the users who become members by non-free methods are users with higher loyalty and stronger spending power, and their consumption needs are relatively rigid. These users are relatively insensitive to prices and pay more attention to consumption efficiency and experience.
In addition, the information asymmetry between the platform and users, the platform displays high-priced products through data analysis, page settings, information hiding, etc., so that users are "killed" unknowingly.
"From the perspective of the regulatory authorities, relevant laws and regulations have yet to be improved, especially for the pricing standards of e-commerce platforms, etc., resulting in a low cost of violations of the platform's big data." Wu Qi said.
"Using algorithm recommendation to customize different information and services for individuals has two risks." Zuo Xiaodong, vice president of China Information Security Research Institute, said that one is the risk of user personal information protection. Platform operators may collect and abuse citizens' personal information in violation of laws and regulations. information; the second is the risk of algorithm manipulation, because the information received and the services used by each person are customized by the operator of the Internet platform, which may lead to public opinion security risks.
Therefore, strengthening the management of algorithm recommendation is not only the need to safeguard the individual rights and interests of citizens, but also the need to safeguard national security.
serious remediation
Promote the healthy development of algorithm recommendation services
In fact, since last year, the regulatory authorities have attached great importance to "big data" and issued relevant regulations.
For example, on April 13, 2021, the State Administration for Market Regulation, together with the Central Cyberspace Administration of China and the State Administration of Taxation, held an administrative guidance meeting for Internet platform enterprises and pointed out that "the implementation of 'big data killing' and other issues must be seriously rectified"; July 2, 2021 , the "Regulations on Administrative Penalties for Price Violations (Revised Draft for Comment)" issued by the State Administration for Market Regulation also stipulates that e-commerce platform operators use "big data to kill familiarity"; The Personal Information Protection Law of the People's Republic of China emphasizes that excessive collection of personal information is prohibited, and businesses are prohibited from "killing big data" through automated decision-making.
It is worth noting that the "Administrative Regulations" issued a few days ago clarified the requirements for the protection of user rights and interests of algorithm recommendation service providers, including guaranteeing the right to know the algorithm, requiring users to be informed of the algorithm recommendation service provided, and publicizing the basic principles and purposes of the service. Intent and main operating mechanism, etc.
It is required that "it is not allowed to use algorithms to implement unreasonable differential treatment and other illegal acts in terms of transaction prices and other transaction conditions based on the characteristics of consumers' preferences and transaction habits."
"This has a greater impact on Internet platform companies in the fields of online sales, life services, social entertainment, information, financial services, computing applications and other fields." Wu Qi said that these companies rely on massive data and algorithms to carry out intelligence for thousands of people. Recommendation, the content of the product recommended by each user is different.
According to the regulations, users have the right to know the algorithm, and the platform needs to inform the user of the algorithm recommendation service, publicize the basic principle, purpose and main operation mechanism of the service, and clarify the corresponding punishment mechanism.
The relevant person in charge of the State Internet Information Office said that the introduction of targeted algorithm recommendation rules and regulations in the field of Internet information services is the need to prevent and resolve security risks, and also to promote the healthy development of algorithm recommendation services and improve the level of supervision capabilities.
Strengthen personal privacy protection
Existing companies are actively deploying
"Securities Daily" reporters noticed that before the "Administrative Regulations" were officially implemented on March 1 this year, some investors "killed the big data" and collected personal information of relevant listed companies on the interactive platform of the Shanghai and Shenzhen Stock Exchanges. Asked about the situation, and some companies responded that they were actively deploying.
An electronic data receipt collection company said that the company has launched law enforcement and enterprise-related products in the aspects of detection and forensics and data security, which are specially designed for dynamic analysis of Android malicious application behavior, and comprehensive real-time analysis of App behavior, such as App reading users. The analysis of information, operation user behavior, etc. provides a fast and expert detection scheme for personal privacy protection.
Li Keshun, president of Guizhou Data Treasure Product Research Institute, believes that for companies driven by algorithms, after the release and implementation of the "Administrative Regulations", users' intervention in the entire service use process will be reduced, and the platform algorithm will gradually adjust to compliance, and some traffic may be It will be lost, but the market competition will be more fair.
An industry insider who provides data services for Internet platforms believes: "The new algorithm management regulations and the personal information protection law can have a deterrent effect on related companies and regulate the healthy operation of the entire industry."
The relevant person in charge of the Cyberspace Administration of China stated that the governance of algorithmic recommendation services requires the joint participation of the government, enterprises, society, netizens and other parties, to promote fairness, fairness, standardization and transparency of algorithmic recommendation services, and to promote the improvement of algorithmic recommendation services to create a clearer network. space.
"Next, regulators, industries and enterprises must form a model of collaborative governance." In Wu Qi's view, regulatory authorities need to improve the policy and regulatory system for collaborative governance, and find out violations of laws and regulations in accordance with laws and regulations; platform companies should effectively perform the main Responsibility, establish and improve the corresponding management system and technical measures.
(Securities Daily)