On "What building an AI-native finance function taught me"

The real lesson here isn't about finance functions—it's that CFOs are finally willing to talk publicly about AI as operational infrastructure, not moonshot. Friar's five lessons probably land somewhere between genuine infrastructure thinking and the polished PR narrative you'd expect from a $80B+ company explaining why it hired expensive talent and overhauled processes. What matters for practitioners is the permission structure: if OpenAI's CFO is naming automated forecasting and AI ROI as table-stakes competencies, your CEO isn't going to accept "we'll figure out AI later" from finance. The shift from "should we use AI" to "how do we build for AI natively" is now expected, not innovative—which means you need to know what native looks like before someone asks why you don't.