Michael Mignano

Post‑AI Firms Win by Openness, Habit, and Scale

Observation: post-AI companies at the app layer have very different strategies around defensibility vs. pre-AI companies. Pre-AI companies focused on user network effects, keeping data and content locked down and proprietary, and doing whatever it took to keep users inside their actual products. Post-AI companies are being forced by the market to make data as open as possible so their users can leverage it as context with their favorite LLM. Collect once, monetize everywhere via APIs/MCP (via usage based pricing, etc). With the latter, the theory of defensibility comes from: 1. Making the use case within the product as habitual as possible. 2. Winning as much marketshare as possible for a specific use case as quickly as possible. The more context you can grab, the more powerful the context becomes, and the more the user would never want to break it apart (even if it's not that hard via APIs etc). 3. Having a good enough product/brand experience such that users won't switch for an experience of similar quality, even if it's offered by the labs and vertically integrated into their models. So, far this has worked to an extent. Or at least it hasn't failed for many. But has it been truly tested yet? APIs and MCP enable labs to ingest all the context they want from as many sources as possible. As they do, the products that generate the context become more and more interchangeable (which is why pre-AI companies favored network effects - to make them stickier and harder to rip out). If you can all of your data out of a product and into the place you actually want to use the data, why do you need that specific product anymore? If the strategy fails, and the labs eat everything, we will look back on the MCP movement as the greatest magic trick the labs ever pulled.