Good morning, you're tuned in to Markets Desk, and here's what's moving the conversation today.
OpenAI's CFO Sarah Friar is pushing back on how companies measure AI investment. Writing in Fortune, Friar argues that seat counts and user adoption metrics are relics of legacy software thinking. What matters now, she says, is the actual work AI completes — outcomes, not logins. It's a framework shift that could reshape how boards justify AI budgets.
Shifting to energy markets, the question over the Strait of Hormuz is no longer hypothetical. The Financial Times is examining how long oil markets can absorb the geopolitical pressure building around that critical chokepoint. Improved oil intensity across major economies has cushioned the blow so far, but analysts warn that cushion may be giving policymakers a false sense of security.
And in AI, markets may be absorbing what some are calling a second DeepSeek moment. Beijing-based Moonshot AI has released Kimi K three, the largest open-weight model ever built, delivering performance close to Anthropic's top model at a fraction of the cost. That efficiency gap is exactly the kind of signal that rattles valuations across the AI infrastructure trade.
That's the tape. Markets Desk, signing off the floor.
