The 'come for the tool, stay for the network' strategy is a foundational model for building successful platforms, as demonstrated by companies like Instagram and Substack.
The current AI era is defined by capital-intensive development, making access to funding a primary competitive moat and fundamentally challenging the traditional, low-cost model of open-source software.
Network effects are evolving; in some modern products, they are 'externalized' to the broader internet ecosystem of influencers and content creators rather than being contained within the product itself.
AI represents a classic disruptive threat to incumbents like Google and will inevitably erode the value of business models built on SEO and referral traffic.
The power of brand and consumer inertia is a significantly underestimated defensive moat in Silicon Valley, capable of protecting a company like OpenAI even if its technical advantages are not unique.
The Open Source Composability Era
Dixon points to the rise of Linux from a hobby project to the world's dominant OS as a prime example of open source succeeding through composability, a model that required developer collaboration but not massive capital.
The Social Network Growth Era
He analyzes how companies like Instagram pioneered growth strategies by first offering a compelling tool (photo filters) to attract users before locking them into a network, and by piggybacking on existing networks like Twitter for initial distribution.
The Great Consolidation Era
Dixon observes that the internet has matured to a state where over 95% of both traffic and revenue are now consolidated in the hands of five to ten large technology companies, setting the stage for new disruptions.
The AI Disruption Era
He identifies the current period as one of classic disruption, where AI-native challengers like OpenAI are threatening the core business models of incumbents like Google, particularly in search.
The Capital-Intensive Moat Era
Dixon argues that the competitive landscape has shifted. In AI, massive capital expenditure has become a primary defensive moat, altering the dynamics for startups and open-source projects compared to previous tech cycles.
▶The Evolution of Network Effects and MoatsApr 2026
Dixon posits that while the classic 'come for the tool, stay for the network' model remains relevant, new forms of competitive advantage are emerging. He theorizes that network effects can be 'externalized' to the broader internet ecosystem of influencers and creators, and that non-technical moats like brand power and consumer inertia are increasingly powerful defenses, as seen with ChatGPT.
Investors should evaluate a company's competitive moat not just by in-product metrics but also by its ability to catalyze an external ecosystem and build a powerful, household-name brand.
▶Capital as the Defining Constraint in the AI EraApr 2026
According to Dixon, the AI sector is fundamentally different from previous software waves due to the massive capital expenditure required to train foundation models. This financial requirement acts as a significant competitive moat, favoring well-funded players and posing a major challenge to traditional, less capital-intensive open-source development.
In the AI space, access to capital and compute resources should be considered a primary indicator of a company's long-term viability, potentially outweighing early user traction or novel algorithms.
▶AI as a Classic Disruptor and Market ConsolidatorApr 2026
Dixon frames the competition between OpenAI and Google as a textbook example of disruptive innovation. He believes AI will inevitably erode business models reliant on SEO and web traffic, while also noting that the broader tech landscape has already seen over 95% of traffic and revenue consolidated among a few large companies.
Analysts should anticipate significant value shifts away from businesses dependent on web referral traffic and toward foundational AI model providers and AI-native applications.
▶The Precarious Future of Open-Source SoftwareApr 2026
Dixon contrasts the historic success of composable open-source projects like Linux with the current challenges facing the movement. He identifies two key threats: the prohibitive capital costs of AI development and regulatory risks, such as proposed liability laws that could create untenable downstream risks for developers and effectively kill open source.
The long-term health of the open-source ecosystem, particularly in AI, is contingent on solving for capital access and navigating an increasingly complex and potentially hostile regulatory environment.