For decades, the equity value in corporations has rested on their ability to build and defend “moats” — structural barriers that protect profits from competition. These moats have come in many forms: scale, brand, distribution, information asymmetry, and switching costs. They created stability for incumbents and concentrated power in large, bureaucratic organizations.

Generative AI is systematically weakening these barriers. The result is not only business model erosion at the firm level but also a potential reshaping of entire economies.

From Scale to Individual Leverage

Scale once justified headcount. Hundreds of analysts, marketers, or designers were needed to sustain output. Now individuals and small teams can harness AI to achieve capacity that rivals entire departments.

• Example: A two-person startup can now produce marketing campaigns, financial models, and investor presentations that previously required dozens of staff. Indie game developers using AI for art and dialogue design can compete with mid-tier studios that once relied on large teams.

Organisational scale still matters in capital-intensive industries (like manufacturing or energy), but in knowledge work it is becoming far less decisive.

From Information Asymmetry to Commoditised Knowledge

Consulting firms, data vendors, and publishers thrived by owning proprietary insights. AI undermines this advantage by synthesizing public data at scale, surfacing answers once locked inside databases or hidden behind paywalls.

• Example: Legal AI tools such as Harvey or Casetext can draft briefs and conduct case law research at a fraction of the cost of junior associates, eroding law firms’ advantage in information-heavy tasks. Market research once requiring paid reports from Gartner or McKinsey can increasingly be replicated with AI-driven synthesis.

Knowledge is becoming less about what you own, and more about how you contextualize and apply it.

From Network Effects to Layer Inversion

Platforms like Amazon, YouTube, and Airbnb have long benefited from network effects. The more users and suppliers they attracted, the stronger their position became.

But AI agents can aggregate across networks, instantly comparing offerings without loyalty to any one platform.

• Example: AI shopping copilots can simultaneously scan Amazon, Walmart, Shopify stores, and independent websites, providing the consumer with a single aggregated recommendation. Similarly, AI travel planners can bypass Airbnb or Booking.com’s curated results, pulling the best lodging options across platforms.

The moat shifts upward, to the AI layer, while traditional platforms risk becoming commoditized suppliers.

From Switching Costs to Frictionless Movement

Banks, enterprise software providers, and telcos have relied on inertia. Once a customer was onboarded, high switching costs ensured long tenure.

AI changes this by automating migrations and transitions.

• Example: Cloud-based AI tools can now port customer data between CRMs like Salesforce, HubSpot, or Zoho with minimal friction. In fintech, account portability, as seen in the UK’s Open Banking framework, allows consumers to shift providers with a single API call.

Especially when regulators encourage it, loyalty is no longer guaranteed.

Productivity, Democratisation, and Risk

This erosion of moats could democratise economies. Bureaucratic drag weakens as independent operators and small firms compete on near-equal footing with incumbents.

• Example: Freelance engineers using GitHub Copilot or Replit Ghostwriter can deliver projects that once required teams in large consultancies. Small e-commerce firms can deploy AI-driven customer service agents indistinguishable from those of multinational retailers.

Value creators — engineers, designers, managers with operating expertise — gain more visibility, unmediated by large institutions. Productivity may rise as barriers fall.

But risks remain:

• Regulated balance sheets (in banking, insurance, etc.) still anchor consumer trust.
• Credibility and reputation remain hard to automate.
• Coordination costs can undermine looser networks of independents.
• Incumbents will lobby to entrench regulatory or compliance barriers.

Policy Implications

If AI is breaking down moats, governments face a choice: preserve incumbent advantages or design policy to harness the productivity gains. Forward-looking policies might include:

1. Encouraging Portability
   • Mandating interoperability and easy switching between providers.
   • Enabling AI-driven consumer choice by ensuring fair access to data and APIs.
2. Supporting Independents and SMEs
   • Providing AI infrastructure credits or public datasets to level the playing field.
   • Reforming procurement to allow small players to compete with incumbents.
3. Rethinking Competition Policy
   • Moving beyond traditional antitrust focused on size, to consider AI-era bottlenecks (such as control over training data or exclusive AI-platform partnerships).
4. Redefining Safety Nets
   • If AI accelerates disintermediation, then traditional employment may fragment. Policy will need to adapt labor protections, benefits portability, and lifelong learning systems.
5. Investing in Trust Infrastructure
   • Governments can play a role in setting standards for authenticity, identity, and fraud prevention and so allowing smaller players to compete with consumer confidence.

The Broader Lesson

Moats have long been treated as synonymous with value. But in the AI era, their collapse may shift value creation toward trust, creativity, execution, and resilience.

The winners will not be those who merely defend barriers, but those who design for a world without them.

For policymakers the challenge is similar: resist the instinct to entrench incumbents and instead build the foundations for a more fluid, democratic, and productive economy.