The AI Organization, part I - by Omar Shams - mutable.ai

My goal is to structure teams around minimizing “coordination headwind,” as described by Alex Komoroske in this deck on seeing organizations as slime mold. The rough idea is that coordination costs (caused by uncertainty and disagreements) increase with scale, and adding managers doesn’t improve things. People’s incentives become misaligned. People... See more

If we want to understand what AIs are going to look like, I think the proto AI that we have are corporations. Corporations are sort of these funny little beasts. They’re not small. I guess they’re not little beasts, but they’re strange. It takes special training to have humans be able to fit within them. They’re made out of humans mostly but they’r... See more

No amount of AI will automate away corporate dysfunction.

In an AI-dominated future, the structure of the American economy is poised for a seismic shift. As AI slashes coordination and transaction costs, the Coase Theorem suggests a fascinating outcome: an economy teeming with smaller, more agile firms. This reduction in costs disrupts the very essence of firm size, allowing nimble players to enter the fr... See more

In a long-context world, maybe the organizations that benefit from AI will not be the ones with the most powerful models, but rather the ones with the most artfully curated contexts.

The two most important problems (at least how I am thinking about them currently), are:

Finding a rigorous scientific framework for how different agent skills, personalities, and instructions combine to be most capable for different problems (think of this as social management science for AI agents) .

Figuring out how you formally validate and verify... See more