One of my favorite Dee Hock frameworks is what he called trying to master the "four ways of looking at things: As they were, as they are, as they might become, and as they ought to be." Through that lens, it's easy to see AI as a technology driving significant change between the way the economy worked five years ago and the way it works today. It's also easy to speculate on where it might lead, especially given the many real risks and challenges around AI alignment. But I'd argue the most important thing anyone can do is develop a perspective on how it "ought to be," and then build toward that. To me, AI shouldn't just be a technology that accelerates the automation of tasks humans used to do. It ought to supercharge our ability to coordinate economic activity, giving people leverage to connect their unique expertise, taste, and data traces with the people and companies that need them. While the AI data labeling ecosystem is one of the first examples of new economic activity with people and companies getting paid by labs for their expertise or data traces, I'd love to see an explosion in AI-enabled freelancers across other sectors of the economy.