Humans don’t speak like forms. A person’s message is vague, incomplete, and sometimes messy: “About that thing from last week, I think we should do something.” To a human familiar with the context, this sentence might mean something; but to a system, it’s a puzzle. We believe the hard job of an intelligent system is exactly this: to pull a right, structured action out of that mess — without asking the user to speak like a form.
Why this is hard
The difficulty comes from the wide gap between human language and a structured action. An action needs clear fields: what, for whom, at what priority. But human language doesn’t state these fields explicitly; it hides them in context, allusion, and assumption. Bridging that gap — not by forcing the user into structure, but by the system understanding the mess — is what makes a good experience.
The path of conversion
This conversion has a few steps. First, finding context: the system has to work out what “that thing from last week” refers to, and retrieve it from shared memory. Second, organising: breaking what was said into meaningful parts — a new fact, a note, a change of state. Third, the stage decision: should this matter advance to the next stage? And finally, action: turning all of this into a clear, structured proposal.
A human in the loop
The important point is that the output of this path isn’t an automatic, final action; it’s a proposal. The system presents what it understood in structured form, and a human reviews and approves it. This loop both catches error and builds trust: the system does the hard work of extraction, and the human remains the final judge.
Domains are only examples
This pattern isn’t tied to one particular domain. It could be a sales note becoming a file update, or a support message reaching a structured item, or a request leading to a specific action. These are only examples; what stays constant is the pattern itself: a vague message on one side, a clear action on the other, and an intelligent path in between.
Putting it together
We believe the mark of a truly intelligent system isn’t answering clear questions, but understanding vague messages. The user shouldn’t have to speak like a machine to be understood; it’s the system that should understand human language. And it’s right here that the gap between a rigid tool and a real colleague shows itself.