Cursor hit 1M active developers, which the company announced with enough specifics — monthly actives, not registered accounts — to take seriously. The growth rate is notable: VS Code took roughly four years to reach comparable penetration in the developer tooling market. Cursor did it in under 18 months.
The reasons matter more than the number. Cursor did not win on features — it won on behavior modification. The product is built around a loop where the model sees your codebase context, you describe what you want, and the model produces diffs you accept or reject. That interaction pattern trains a new habit: thinking in terms of intent rather than implementation. Once developers internalize that habit, switching back to standard autocomplete feels like a regression.
The adoption curve tells you something specific about AI-native products: they don’t spread through procurement decisions the way enterprise software does. They spread through individual developers trying them, getting hooked, and then either bringing them into their team or getting frustrated when their team doesn’t use them. The bottom-up growth pattern is critical — it means Cursor didn’t have to win a single procurement process to get to 1M developers.
The strategic risk Cursor faces at scale: GitHub Copilot has distribution that Cursor cannot match through organic growth alone. Microsoft can bundle Copilot into GitHub Enterprise at a price point that changes the procurement calculation even for teams already using Cursor. The next 12 months will reveal whether Cursor can build enough organizational integration (code review, CI, team collaboration) to create switching costs that survive Microsoft’s pricing flexibility.
The note on the acquisition by the instruction-writers: Cursor acquiring Windsurf in early 2026 is the response to that pressure.