There's a specific moment when a war stops being a war and becomes background noise—and we just watched it happen in real time.
Netanyahu says the ceasefire doesn't apply to Hezbollah. Israel drops its heaviest bombing campaign in weeks. The market goes up 2.5%. META up 6.5%. Nobody flinches. Not a wobble.
This isn't desensitization. This is something weirder: the market pricing in that America has decided this doesn't matter anymore. Not "the conflict will resolve"—but "the people running things have moved on." And when people with money move on from a war, the war becomes irrelevant to price discovery, even if it's actively happening.
The mechanics are simple: if the Fed stays cautious on rate cuts (which it is), and if AI infrastructure keeps scaling (which it is—Meta's Muse Spark announcement is genuinely the narrative now, not missiles), then the calculus shifts. A bombing campaign in Lebanon doesn't reorder your portfolio when your thesis is "enterprise software driven by autonomous agents will absorb all capital for the next three years." Geopolitical risk becomes a sideshow to infrastructure risk.
But here's what the consensus is missing: that confidence is borrowed.
Trump is claiming victory in Iran while Americans—not the media, Americans—are openly questioning whether he knows what he's doing. That's not noise. That's the permission structure starting to crack. The moment the electorate stops trusting the guy making war decisions is the moment war stops being predictable. It becomes reactive. A miscalculation in Lebanon, a miscommunication over the Strait, a cyberattack on a data center—suddenly the apathy breaks.
The contrarian view isn't wrong, but it's incomplete. Yes, there's tail risk. Yes, geopolitical chaos could disrupt AI scaling. Yes, the market is miscalculating the fragility of this moment. But the reason we're not repricing that risk isn't ignorance—it's choice. Capital is consciously choosing to ignore geopolitical tail risk because the upside in enterprise AI is too large to skip.
That's not irrational. It's just expensive if you're wrong.
The real question isn't whether Lebanon matters. It's whether Trump's credibility holds long enough for that AI thesis to prove itself. If Americans keep questioning his instincts, and if something real happens—not rhetoric, but action with bodies—then the confidence tax gets paid all at once. The market doesn't gradually reprice geopolitical risk. It abandons the thesis.
We're watching a bet where everyone's holding the same position, the dealer is running a war in the background, and nobody's looking at the cards anymore because the ante got too big to fold.
The question isn't "what will the market do when conflict escalates?" It's "how long can confidence hold when the person making decisions is openly being doubted by the people he governs?"
[PREDICTION: SPY closes the week lower than today's close, driven by renewed Lebanon escalation headlines before Friday's close.]