Alphabet’s AI commitments pass $1tn, its quarterly filing shows
THE SO WHAT
Alphabet quietly committing over $1T to future compute, energy, and data centers—off balance sheet—confirms that AI infra is being treated like multi-decade utility buildout, not cloud opex. If your product depends on their APIs or ad rails, assume they will prioritize AI-native monetization and cross-subsidize aggressively from this capex base.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIMicrosoft made $10bn from OpenAI last quarter, 11% of its revenue
When 11% of Microsoft’s quarterly revenue is tied to OpenAI, AI stops being a side bet and becomes a core dependency—on both sides. If you’re building on this stack, vendor concentration risk is now macro-scale, not just a line-item concern.
Applied AIAI Creating New Pathways to Engage Clients, Says Schwab CEO
When a firm like Schwab is publicly crediting AI with new client engagement channels and backing startups like Nevis, “wait and see” in financial services becomes a risk posture. If you own a client-facing P&L, assume your peers are already testing AI-native touchpoints and measure against that bar.
Applied AIMeta quietly launches an AI fashion app called Style Sketch
Meta wiring an AI fashion app directly into Instagram closets is another move to keep shopping, styling, and data exhaust inside its graph. If you’re in ecommerce or fashion, plan for discovery and conversion to keep shifting into closed, AI-personalized surfaces you don’t fully control.
Applied AIBaseten’s joey-1 makes AI decisions for a penny per million tokens
A 2B open-weights decision model at $0.01 per million tokens and ~20ms latency is a shot at replacing full LLM calls for routing, ranking, and simple judgments. If your stack still uses frontier models for these primitives, there’s now a clear cost and latency wedge to exploit.