AI won't save a product that was never composable Most product teams shipping AI right now are following roughly the same playbook. A chat sidebar. A summariser. A copilot panel that can answer questions about whatever's on screen. Each release gets a launch post. Each one tests reasonably well in isolation. And yet, a year in, something is off. The product feels louder, not better. Customers use the AI features once or twice and go back to doing things the old way. The team keeps shipping. The needle doesn't move. This is not an unusual story. It's close to the median story. And the diagnosis most teams reach for, that the prompts need tuning or the model needs upgrading or the UX needs another pass, is almost always wrong. The problem is further upstream, and it's structural. The product wasn't built to be recomposed. AI can only ever sit on top of it. And AI on top of a rigid product is a feature, not a transformation. AI won't save a product that was never composable.…