The Morning in Review

Tuesday's session opened with a wave of AI-related dealmaking that would have looked extraordinary even six months ago. The headline was the Google-Blackstone joint venture: Alphabet (GOOGL) and Blackstone are forming a new AI cloud company with $5 billion in initial committed capital and a potential total commitment of $25 billion. Blackstone is expected to hold a majority stake. The venture is explicitly positioned to compete with CoreWeave, the GPU-focused cloud provider that went public earlier this year, and to challenge Nvidia's (NVDA) grip on AI infrastructure supply chains.

The deal is notable not just for its scale but for what it represents structurally. Alphabet gains AI compute capacity without bearing the full balance-sheet burden. Blackstone deploys institutional capital into a category it has been building exposure to across its real estate and infrastructure funds. Together they create a well-capitalized alternative to the hyperscaler-or-Nvidia binary that has defined AI infrastructure to this point.

Elsewhere, a jury dismissed Elon Musk's lawsuit against OpenAI and Microsoft (MSFT), clearing a legal obstacle that had complicated OpenAI's corporate restructuring and its path toward an IPO. Microsoft, as OpenAI's largest backer, is the primary beneficiary of that verdict. The timing of any OpenAI listing has not been confirmed, but the legal clearance is a prerequisite for the process to move forward.

The Tension Beneath the Headlines

Not everything in the morning session pointed in the same direction. Bank of America's May Global Fund Manager Survey — released earlier this month but drawing fresh attention today — found that 73% of professional investors now identify long global semiconductors as the most crowded trade on Wall Street, a sharp rise from April's reading. The names most associated with that trade: NVDA, AMD, Micron (MU), and Intel (INTC).

Crowded-trade readings matter because they signal asymmetric downside risk. When the vast majority of professional capital is already positioned on one side, the marginal buyer is scarce — and even a modest negative catalyst can trigger disproportionate selling as funds race to reduce exposure simultaneously. The survey does not predict a reversal, but it narrows the margin of safety for anyone adding to semiconductor exposure at current levels.

The irony is hard to miss: the same session that produced a $5 billion AI infrastructure commitment is also flashing the clearest crowding warning the semiconductor trade has seen in months. Both things can be true — fundamentals can be strong while positioning is stretched — but the combination raises the stakes for any near-term catalyst that disappoints.

The Macro Layer

Geopolitical risk is adding another complication. Moody's Analytics' chief economist warned this week that escalating tensions involving Iran could produce what they described as "unmoored" inflation — price pressures that become difficult for the Federal Reserve to anchor with conventional tools. Strategist Ed Yardeni went further, flagging a potential Fed rate hike as early as July if inflationary pressures intensify.

Energy prices are the primary transmission mechanism. Iran is a significant oil producer, and any sustained conflict scenario could push crude sharply higher, feeding directly into headline consumer price inflation. The Fed has held rates steady in recent months after an extended tightening cycle; a resumption of hikes would represent a meaningful policy shift and would pressure equity valuations — particularly in rate-sensitive and high-multiple sectors like AI and semiconductors.

Markets have not yet fully priced this scenario. That gap between current pricing and a plausible tail risk is exactly the kind of asymmetry that tends to matter when positioning is already stretched.

The Utility Signal Worth Watching

NextEra Energy's announced acquisition of Dominion Energy in an all-stock deal valued at approximately $67 billion is the session's most important non-tech story — and it connects directly to the AI infrastructure theme. NextEra, already the world's largest producer of wind and solar energy, is acquiring Dominion at a moment when electricity demand forecasts are being revised sharply upward, driven in part by the power requirements of AI data centers.

The deal structure — all-stock, meaning Dominion shareholders receive NextEra shares rather than cash — avoids large debt loads but leaves both sets of shareholders exposed to market fluctuations before regulatory approval. A transaction of this size will face a lengthy review process. What the deal signals, regardless of its eventual outcome, is that utility-scale energy infrastructure has become strategically critical in ways that are attracting capital at a scale previously reserved for technology M&A.

For investors tracking the AI infrastructure buildout, the NextEra-Dominion deal is a reminder that the demand story extends well beyond chip designers and cloud providers. Power generation, transmission, and storage are now integral parts of the same capital cycle.

Afternoon Setup

The afternoon session carries a specific set of risks worth monitoring. Semiconductor names face the dual pressure of stretched positioning and a macro backdrop that is incrementally less favorable than it appeared a week ago. Any further commentary on Iran, oil prices, or Fed policy expectations could move rates markets and ripple into high-multiple equities.

The OpenAI IPO verdict clears a legal overhang for MSFT, but the stock's near-term reaction will depend more on broader market tone than on the legal outcome alone. The Google-Blackstone venture has no immediate earnings impact for GOOGL — the capital commitment is large but the venture is in early stages.

Walmart (WMT) is scheduled to report first-quarter earnings on May 21, which will offer a read on consumer spending health and whether tariff-related cost pressures are being passed through to shoppers. Nvidia's results, expected later this week, remain the single most anticipated catalyst for semiconductor and AI-exposed equities — and will be parsed closely against the crowding warning as a test of whether the fundamental case still justifies the positioning.