Skip to main content

I'm hearing we have the best words.

From our brains to your screen, get the latest from our team on digital strategy, building better teams, and making the most of AI.

September 14, 2026

The Agent Door Test

As we think about handing more work off to AI agents, we need a framework for understanding both the type of work we're delegating and the consequences when an agent gets something wrong.

The one-way versus two-way door distinction gives us a useful mental model for what kind of work agents can do. It helps us determine the level of autonomy they should have.

May 25, 2026

When AI Fixes the Summary but Leaves the System Wrong

Mostly Serious tested AI-assisted workflows in Vei, our internal test environment for realistic business systems. In 32 of 36 completed runs, at least one field we checked was still wrong after better information arrived.

The issue was not always that the AI misunderstood the situation. Often, the AI made the summary better while failing to update the fields another person would rely on later.