Amadeus reinvests heavily in R&D, with a spend equivalent to its #3 competitor's total revenue. This creates a widening technology and product gap that smaller players cannot bridge, fortifying its market leadership and making it increasingly difficult for others to keep up.

Incumbents improve exponentially (e.g., 2x annually). A startup taking 5 years to reach scale needs a 100x better product at launch just to be competitive, accounting for the incumbent's 32x improvement during that development time.

The AI company with the largest share of coding-related tokens may gain an insurmountable lead. More developers using the tool generate more training data and access to codebases, which in turn improves the model's capabilities, creating a self-reinforcing cycle that consolidates market dominance.

Contrary to the belief that number two players can be viable, most tech markets are winner-take-all. The market leader captures the vast majority of economic value, making investments in second or third-place companies extremely risky.

The moat for a market leader isn't just the initial VC investment; it's the subsequent, rapid follow-on rounds that create a 'wall of money.' This forces competitors to prove they can win against not just a brand name, but also a massive and compounding capital advantage.

Unlike traditional startups where excess capital creates bloat, frontier AI companies can convert dollars directly into compute. This immediately improves the product and reinforces the market leader's competitive advantage, fundamentally altering the scaling dynamics for venture-backed companies.

For decades, you couldn't catch a competitor with a two-year lead just by hiring more engineers. AI changes this. Access to massive capital for compute (GPUs) and data now allows teams to solve problems and close gaps quickly, making capital itself a primary competitive moat.

The true competitive differentiator in the AI era won't just be adoption speed, but how companies reinvest efficiency gains. Leaders will funnel savings back into AI innovation, creating a compounding effect that leaves laggards permanently behind, making stock buybacks an expensive choice.

Unlike pure software, the value in physical AI and hard tech comes from long-term compounding of technology. Startups often fail because they don't survive long enough to see these returns. This makes early commercial discipline and constraints crucial for longevity.