算法“杀熟”需多元共治

Core Viewpoint - The evolution of algorithm technology has transformed the phenomenon of "big data price discrimination" from overt to covert forms, shifting from explicit price differences based on static identities to implicit systemic games relying on artificial intelligence and diverse contexts [1][3] Group 1: Evolution of Price Discrimination - The early stage of "price discrimination" was characterized by easily identifiable "binary pricing," where platforms set different prices for "new users" and "old users," leading to higher prices for loyal customers [3][4] - With advancements in algorithm technology, pricing mechanisms have fundamentally changed, moving from static identity profiles to dynamic assessments of consumer "real-time situations" and "instant payment willingness" [3][5] - Algorithms can now capture multiple real-time variables, such as device battery life and geographical location, to reconstruct an individual's "price elasticity" and detect their maximum willingness to pay [3][5] Group 2: Mechanisms of Price Discrimination - Leading online travel agencies (OTAs) have developed a multidimensional implicit price discrimination system, including hardware-based discrimination, where users with high-value devices see hotel prices 8% to 15% higher than those using ordinary devices [4][5] - Dynamic pricing based on "search anxiety" is implemented when algorithms detect high-frequency searches by a user, leading to price increases, while new accounts see the initial price [4][5] - The strategy has evolved from overt price adjustments to covert "shadow pricing," embedding differential operations in non-price dimensions like coupon distribution and search result rankings [5][6] Group 3: Economic Implications - The prevalence of "big data price discrimination" is directly related to the development stage of the platform economy, transitioning from "expansion" during the traffic dividend period to "extraction" of existing user value in the stock competition phase [8][9] - This "stock harvesting" model creates an internal paradox, as platforms should lower information costs to create value, but instead, they increase transaction friction through personalized information barriers [8][9] - The algorithmic pricing mechanism distorts the core function of the market—price signals—leading to a loss of market vitality and potential collusion among platforms [8][9] Group 4: Governance and Regulation - Governance must move beyond punitive measures to a systemic reconstruction that includes legal regulation, technical audits, market structure optimization, and social supervision [9][10] - Legal rules should be refined to enhance operability, introducing a "reversed burden of proof" mechanism to hold platforms accountable for price discrimination [10][11] - A multi-stakeholder governance ecosystem should be established, including independent third-party algorithm auditing institutions and consumer empowerment tools to enhance transparency and consumer rights [12][13]