The promise of widespread enterprise AI is held back by a fundamental problem: many companies still run on legacy, on-premise systems from the 80s and 90s. This "digital transformation" bottleneck must be solved first, as AI can't be adopted until the prerequisite move to modern cloud infrastructure is complete.

Existing companies ("AI emergent") are structurally disadvantaged by legacy tech, talent resistant to change, and outdated pricing models. AI-native startups, built from the ground up with AI, hold a significant advantage that even giants like Apple struggle to overcome.

Established SaaS companies struggle to implement AI because their teams are burdened with supporting existing customers, fixing feature gaps, and fighting legacy competitors. AI-native startups have a massive advantage as they don't have this baggage and can focus entirely on the new paradigm.

AI coding's true enterprise value is limited because models struggle with legacy systems. Companies run on trillions of lines of mediocre code in old languages like COBOL—a problem that requires human intervention over decades, not a simple AI solution, which limits immediate, real-world impact.

The rapid evolution of AI makes it difficult for established startups with existing teams and processes to adapt. It can be trickier for a company with "legacy stuff" to pivot its workforce and culture than for a new, agile founder starting with a clean slate.

Incumbents face the innovator's dilemma; they can't afford to scrap existing infrastructure for AI. Startups can build "AI-native" from a clean sheet, creating a fundamental advantage that legacy players can't replicate by just bolting on features.

Large companies like Google and Meta must undergo a painful process of reinventing their "classic consumer software building factory" for the AI era. Startups have a key advantage: they can build AI-native processes and cultures from a blank slate, which is often easier than retrofitting a massive organization.

AI coding assistants have recently crossed a critical threshold. They are no longer just for building new features but are now highly effective at refactoring legacy code. This dramatically changes the economics of modernizing established software companies by accelerating the notoriously slow process of paying down technical debt.

To transition to AI, leaders must ruthlessly dismantle parts of their existing, money-making codebase that are not competitively differentiating or slow down AI development. This requires overcoming the team's justifiable pride and emotional attachment to legacy systems they built.