The Impact of Generative AI on Content Platforms: A Two-Sided Market Analysis with Multi-Dimensional Quality Heterogeneity

TL;DR

The paper addresses how Generative AI reshapes welfare, inequality, and diversity on two-sided content platforms by modeling the co-evolution of AI, human creators, and consumers in a multi-dimensional quality space. It develops a unified framework that endogenizes AI learning and human adaptation, and introduces a Policy Trilemma across allocative efficiency, distributional equity, and ecosystem sustainability, analyzed through a micro-founded static benchmark and agent-based simulations. The demand side is captured by a heterogeneous Mixed Multinomial Logit model, while humans incur convex costs and AI has near-zero marginal costs, enabling a realistic exploration of competition, displacement, and welfare shifts. Key findings show that AI improves consumer surplus and lowers prices but increases concentration and displaces human talent, while governance regimes that promote creativity and diversity (e.g., Pro-Creative, Low-Commission) can achieve Pareto-efficient outcomes that preserve the long-tail and broad welfare gains, highlighting a practical path for policy and platform design in the GenAI era.

Abstract

This paper presents a unified computational framework to examine how generative AI (GenAI) reshapes welfare, inequality, and diversity in content platform economies. By integrating welfare economics with agent-based simulations, we model the co-evolutionary dynamics among AI generators, human creators, and consumers within a two-sided market characterized by multi-dimensional quality heterogeneity. Unlike static models, our framework endogenizes AI learning as a function of human data synthesis and models human adaptation as a strategic reallocation of skills toward creative niches. The results reveal that while GenAI significantly enhances consumer surplus through technical quality gains and price depression, it triggers a skill-biased displacement of human incumbents and intensifies market concentration. Through the evaluation of six governance regimes, we identify a fundamental ``Policy Trilemma'' where platforms must navigate non-trivial trade-offs between allocative efficiency, distributional equity, and ecosystem sustainability. Our findings highlight that algorithmic diversity and pro-creative commission structures function as essential economic mechanisms for sustaining long-tail participation and inclusive social welfare in the generative AI era.

Figures (7)

• Figure 1: The Co-Evolutionary Mechanism of AI and Human Creators. This system dynamics diagram illustrates the feedback loops between human supply ($S_H$), training corpus diversity, and AI quality learning. The red loop highlights the risk of Model Collapse: as AI price advantages displace human creators ($N_H \downarrow$), the diversity of the training corpus diminishes, leading to stagnation in AI quality improvement.A cycle diagram showing the feedback loop between human creators and AI. Blue boxes represent human factors like Supply and Skill. Red boxes represent AI factors like Quality and Price. Arrows indicate positive or negative feedback. The center shows a warning sign for Model Collapse Risk.
• Figure 2: Market Segmentation in Multi-Dimensional Quality Space. The phase diagram visualizes the separation of consumer preferences based on the threshold $\theta^* = \beta_t / \beta_c$. The orange line represents the equilibrium threshold. The blue trajectory shows the Creative Escape strategy, where human creators (blue circle) actively shift their skill vector towards the creativity-intensive region ($q_c \uparrow$) to avoid direct competition with the technically dominant AI (red square).A 2D phase plot with Technical Quality on the X-axis and Creative Quality on the Y-axis. A diagonal orange line separates the AI Dominance zone from the Human Refuge zone. Arrows show the movement of Human and AI agents over time, illustrating skill adaptation
• Figure 3: Baseline Simulation Dynamics (H1--H6). Panels confirm asymmetric AI learning (H1), skill-biased human adaptation (H3), and the S-shaped adoption of AI (H4).
• Figure 4: Welfare Decomposition Over Time: Demonstrating the redistribution of surplus from human producers to consumers post-AI entry.
• Figure 5: End-State Summary Heatmap: Policy vs. Metric performance comparison.

Theorems & Definitions (15)

• Proposition 4.1: Equilibrium Price Ceiling
• Proposition 4.2: Welfare Decomposition and Displacement
• proof
• proof
• Lemma B.1
• Theorem C.1: Existence of Equilibrium
• proof
• Lemma C.1
• Lemma C.2
• Lemma C.3