Expansion of AI Companies into Applications
First, major AI model providers increasingly expand into application markets to recover massive infrastructure costs required to develop and operate large AI systems.
Consequently, these companies begin competing directly with their own customers, particularly startups and software firms that build applications using their AI models.
As a result, AI platforms simultaneously serve as suppliers of core technology and as rivals in the same downstream markets.
Market Structure of AI Models
Meanwhile, the market for general-purpose large language models increasingly resembles a commodity industry, where capabilities converge and differentiation becomes difficult.
In addition, rapid imitation allows well-funded competitors to replicate innovations quickly, intensifying rivalry among leading AI developers.
Currently, the industry resembles an oligopoly dominated mainly by companies such as Google, OpenAI, and Anthropic.
Together, these firms controlled roughly ninety percent of the enterprise AI model market by the end of 2025.
Conflict of Interest in Platform Competition
However, vertical integration between AI platforms and application developers generates a structural conflict of interest.
Specifically, companies controlling essential AI models also compete against businesses that rely on those models to build competing services.
Therefore, the platform operator may have incentives to favor its own applications while disadvantaging independent developers.
Risks of Discriminatory Platform Behavior
In practice, AI providers could degrade service quality, restrict access, or deny resources to competing developers.
At the same time, the platform may grant itself preferential access to data, computing resources, or model capabilities.
Consequently, competing application developers risk losing the technological foundation necessary to run their businesses.
Limits of Current Antitrust Policy
Although these risks resemble past disputes involving digital platforms, existing antitrust laws may struggle to address the problem effectively.
Furthermore, political conditions make new regulatory protections unlikely in the near future.
Nevertheless, policies preventing unreasonable discrimination among similarly situated customers could reduce conflicts and support continued innovation.
Source:
Brookings Institution. (2025). What happens when AI companies compete with their customers?https://www.brookings.edu/articles/what-happens-when-ai-companies-compete-with-their-customers/