AI May Redefine What Makes Platforms Competitive

Watch more: TechREG: The Wharton School’s Pinar Yildirim

Network effects have long powered digital platform growth, creating a flywheel where each additional user makes a service more valuable and draws in more users.

That dynamic shaped the rise of social networks, marketplaces, ride-hailing apps and streaming platforms. However, as artificial intelligence embeds itself across the digital economy, that familiar loop is beginning to shift.

That was the focus of a TechREG interview with Pinar Yildirim, associate professor of marketing and economics at The Wharton School of the University of Pennsylvania. Speaking with Competition Policy International (CPI), a PYMNTS company, Yildirim said AI is not only strengthening certain network effects but also substituting for them in ways that weaken the long-standing link between user scale and market power.

Yildirim also examined these dynamics in a paper for a CPI Antitrust Chronicle on AI and network effects.

“The reliance on network effects in order to grow might be less and less of a concern as AI starts to replace some of the functions that network effects have been serving,” she said.

A shift is happening in how platforms shape user experiences, Yildirim said. Traditional recommendation systems learned from the behavior of large groups. What many people watched, bought or listened to informed what individuals discovered.

AI-driven personalization changes that. Modern recommendation systems build a different version of the platform for each user. Home screens, rankings, playlists and content can differ between two people on the same service.

“We are almost instinctively consuming a different version of Spotify, consuming a different version of Netflix,” Yildirim said.

That fragmentation potentially lowers the value created by more users. With fewer shared touch points, the value added by additional participation declines.

“That definitely reduces the reliance on network effects in order to keep the consumption going,” she said.

AI also changes competitive dynamics by lowering the marginal cost of producing text, code, audio and video. Tasks that once required teams of writers, designers or developers can now be automated or accelerated.

This reduces a traditional advantage of large platforms. Scale once helped companies produce and distribute high-quality content. As those costs fall, smaller producers can enter markets with content that looks and feels professional.

“To the extent that we lower the costs of production, we also lower the cost of entry into different areas,” Yildirim said.

AI “should increase the ability of the long-tail producers to enter a market,” she added.

That shift weakens the degree to which user scale translates into lasting market power.

AI is also substituting for user-generated content. Platforms like YouTube, Facebook, Reddit and Quora rely on users posting, answering questions and interacting. Traditionally, more users meant more content and engagement.

Generative models weaken this dependency. AI tools can answer questions, write posts and summarize thousands of reviews, reducing the need for human input. Platforms can increasingly rely on “the wisdom of the internet” embedded in large language models when human contributions are limited, Yildirim said.

If algorithms supply or compress much of the content, each new user’s contribution matters less. The traditional flywheel stops working the way it used to.

Yildirim said she believes the assumption that a big user base indicates power in the market no longer holds.

“We cannot necessarily say a platform with a smaller installed user base is not going to have competitive power,” she said.

Algorithms can now provide advantages that historically came from scale, making smaller platforms more competitive than their size would suggest.

“It’s quite possible that algorithms can now start to substitute for those network effects,” she said.

AI also challenges long-standing economic frameworks. For years, multi-sided market theory described how platforms create value by connecting different groups of users. AI alters that logic by reducing reliance on human-to-human interaction.

Economists and policymakers will need to “update our prior knowledge” because competitive power now has multiple sources, Yildirim said. User contributions, data-driven effects and algorithmic performance all shape outcomes.

Understanding these forces will require more research. Studying how much AI substitutes for user-generated content, how personalization fragments audiences, and when AI amplifies rather than diminishes network effects is important, she said.

Network effects still matter, but not in the same way. Algorithms now perform functions that previously relied on large user communities. That shift means scale is no longer a reliable indicator of competitive strength.

For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.

The post AI May Redefine What Makes Platforms Competitive appeared first on PYMNTS.com.