Insights
The decisions made
before the work begins
Enterprise AI does not usually fail at deployment. It fails earlier, in decisions made before the work begins — and those decisions are rarely recognized as decisions at the time.
Most institutions can tell you whether an AI program worked. Far fewer can tell you whether the institution is worth more because it happened. Those are different questions, and it turns out only one of them scales.
The work collected here examines those moments across different subjects: the early decisions and hidden assumptions that determine whether transformation earns a place in the future.
The papers
Whitepaper
The Foundations Framework
A governing standard for AI transformation value creation in regulated industries
A pilot can work, a business case can hold, a team can perform — and the program can still fail to scale. The paper sets out the five layers that govern whether an AI program creates durable institutional value — and the standard the next dollar of capital is committed against.
Is the institution worth more because the program happened?