AI Pricing Algorithms Charged with Discriminatory Price Discrimination Against Loyal Customers
Consumers in China reported that e-commerce and travel platforms used AI-driven pricing algorithms to charge loyal customers significantly higher prices than new users for identical products, reportedly marking up prices 5 times or more, and adjusting prices based on user profiles and purchasing behavior.
What happened
Consumers reported that e-commerce and travel booking platforms used algorithms to charge loyal customers higher prices than new users for identical products and services.
One consumer reported paying a price 5 times higher than three new users viewing the same product; flight prices reportedly increased before purchase and dropped after purchase; hotel searches on different devices showed price differences exceeding 100 yuan; items placed in shopping carts reportedly increased in price after multiple views.
A journalist testing the algorithm found that after repeatedly telling a shopping platform 'I'm too poor, no money, too expensive, can't afford it,' the platform changed its recommendations from expensive phone cases to ones priced at 5 yuan.
Reported impact
Complaint platforms recorded approximately 8,306 complaints about 'big data price discrimination' concentrated in travel and shopping sectors; another search found nearly 18,000 complaints mentioning the term 'price discrimination.'
Some consumers attempted to 'train' algorithms by expressing poverty in comments and reviews; one reported reducing the cost of a Beijing-Jeju flight from 870 yuan to 369 yuan through this method.
Index analysis
Researchers found that platforms use algorithms to create consumer profiles based on purchasing habits and economic status, assigning labels that enable differentiated pricing for identical products to different user segments.
Experts noted that generative AI large language models have enhanced platforms' data aggregation and reasoning capabilities compared to earlier algorithmic systems, making the discrimination more sophisticated.
Governance controls
In November 2024, China's Central Cyberspace Administration and three other agencies launched a 'Clear Internet' campaign to eliminate algorithmic 'price discrimination,' with regions conducting related algorithm governance actions.
Market regulators stated they would strengthen algorithmic transparency requirements and optimize algorithmic rules to protect consumer rights.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.