February 23, 2026
Google's WebMCP: The Birth of the AI-Friendly Product
WebMCP proposes something structural: allowing sites to expose their capabilities declaratively to AI models. Products that are more legible to machines tend, paradoxically, to become clearer to humans as well.
For years we talked about user-friendly products, mobile-first, accessibility-first, data-driven. It seemed like we had a fairly stable checklist for digital maturity.
That checklist may be becoming incomplete. WebMCP, an evolving initiative from Google, proposes something structural: allowing sites to expose their capabilities, states, and actions declaratively to AI models. Rather than relying exclusively on the visual interface and heuristic interpretation, agents could interact with systems based on formalized context.
It is still early. It may gain traction, change significantly, or simply never consolidate. That is part of the industry's innovation cycle. But regardless of where WebMCP specifically lands, the discussion it raises is hard to ignore.
The web went through a similar transformation when accessibility stopped being an optional detail and started demanding consistent semantics, clear structure, and explicit intent in components. Teams that treated it purely as compliance did the minimum required. Teams that understood it as an architectural discipline ended up strengthening their design system, their clarity of state, and their product coherence.
The idea of AI-friendly could follow the same path. Not as an extra layer to please models, but as an incentive to make explicit what the system does, what state it is in, and what actions it actually performs. Products that are more legible to machines tend, paradoxically, to become clearer to humans as well.
For those working in product, the question is beginning to change shape: is your system merely usable by people, or is it structurally operable?
User-friendly remains non-negotiable. But we may be entering a phase where designing only for humans is no longer enough.