AI Dominates Venture Capital: 50% of May-June Funding Focused on Machine Learning

1,335 venture funding announcements in thirty days. $2.3 trillion in stated valuations and capital commitments. The sheer volume alone signals something fundamental shifting in how technology capital flows. But the real story is narrower and more acute: artificial intelligence is pulling nearly everything else to the sidelines.

Half of all announced deals this month explicitly mention AI or machine learning. That's not incremental growth. That's market gravity reverting.

The AI Supercycle Is Real

May and June 2026 are not outlier months in venture funding. They are windows into the current structure of the market. When Anthropic announces a $65 billion Series H, when Alphabet commits $80 billion to AI infrastructure, when a stealth-mode startup like Prometheus emerges with a $41 billion valuation—these are not anomalies. These are the shape of capital allocation right now.

The numbers defy intuition. Anthropic's valuation of nearly $1 trillion would place it among the five largest public companies on earth, yet it remains private. SpaceX, similarly valued, has not completed an IPO despite fifteen years of operation. The distinction between "Series" and "valuation" in these mega-rounds is becoming semantic: the announced figures for Anthropic, SpaceX, and Switch reflect investor confidence in future multiples, not current round pricing.

What matters: the capital is deployed. Anthropic is hiring, building, training models. Alphabet is purchasing compute. Switch is expanding capacity. The valuations are real proxies for how the market is pricing AI infrastructure and capability in 2026.

Healthcare, fintech, B2B software—sectors that dominated venture nine months ago—now represent 5% of announced deal volume. Not zero. Five percent.

Stage Structure Is Collapsing Into Mega-Rounds

Three-quarters of the funding announcements tracked do not specify a traditional Series stage (A, B, C, etc.). Most are described as "valuations," "fundraisings," or "strategic investments." This is a statistical artifact of the data: investors and founders are no longer creating named tranches. They are closing growth rounds without stage labels, or they are announcing internal valuations without a corresponding external round.

Among deals where round stage is explicit: Seed and pre-seed rounds account for 129 announcements (10%). Series A, 84 (6%). Series B and later, fewer than 100 combined (2%). The modal deal in the market is either seed-stage or uncharacterized. The median round size, when identifiable, is $21 million. The mean is $3.8 billion—a gap that reflects a handful of mega-rounds weighing heavily on the average.

This distribution has practical implications. Seed-stage investors remain active. But the gravitational center of capital is moving past traditional Series stages entirely. Mega-rounds in AI infrastructure are a different asset class: they are quasi-VC, quasi-PE, quasi-strategic. LPs are investing alongside corporate arms, alongside growth funds, alongside others taking fundamentally different return profiles.

Geography and Sector Are Secondary

In typical analysis, we would break down the 1,335 deals by geography—US, EU, Asia-Pacific—to understand regional capital flows. The data exists. But it serves less explanatory power than it once did. What matters is whether a company is positioned in AI. Geography is noise by comparison.

The same applies to "fintech" or "healthcare" or "real estate tech." These categories still receive capital. But they are funded at traditional venture scales: $5M–$20M seed rounds, $20M–$50M Series A. In a market where the median announced deal is $21M, startups in non-AI sectors are indistinguishable from those from 2021 or 2015 in terms of capital velocity.

The bifurcation is now clear: one market for AI (unrestricted capital, billion-dollar+ expectations, twelve-month fundraising cycles), and another market for everything else (traditional early-stage economics, greater capital discipline, exit visibility within 5–7 years).

What Happens to Returns?

Investors and founders are pricing Anthropic at $1 trillion. OpenAI is in discussions that imply $100 billion+. xAI, less than two years old, is reported at $20 billion. Valuations are not forecasts—they are collective bets. The implied message: whoever wins the AI race will be worth more than Microsoft, Google, or Saudi Aramco.

Mathematically, this is dicey. No venture firm can afford a $100B+ loss on a mega-round, so the success case is assumed. But distribution of outcomes in AI is not normal. Winners take most. The probability-weighted return on capital deployed to Anthropic, xAI, or DeepSeek is either enormous or catastrophic. There is no middle ground for companies worth $500B+ before profitable operations.

For LPs: this is the cost of participation in the category. For traditional venture (sub-$100M fund sizes), this is a problem. The largest deals are now off-market or closed to smaller firms. Capital concentration is increasing.

What This Means for the Next Thirty Days

If the pattern holds, June will show continued AI dominance but flattening growth—the 409 deals reported in the final week of June are anomalously high, likely reflecting end-of-quarter announcements that were already in flight. Expect the weekly rate to normalize to 200–250 by early July.

Watch for: (1) traditional venture funds announcing new AI-focused vehicles (this has already begun); (2) corporate VC arms closing AI infrastructure deals larger than the LP base of most venture funds; (3) seed rounds in AI infrastructure—data pipeline tools, model inference engines, fine-tuning platforms—as LPs notice the pick-and-shovel opportunity behind the mega-rounds.

The venture market is not broken. It is reorganizing. For founders outside AI, this is a harder fundraising environment than 2021 or 2022, but capital has not frozen—it has migrated. For VCs, the question is no longer "can we find good deals?" It is "can we participate in the AI mega-round ecosystem?" The answer, for most, is no.