AI Funding Dominates Global Venture Capital — $2.5 Trillion in 30 Days Shows Where Capital Is Really Flowing

Venture capitalists deployed roughly $2.5 trillion in artificial intelligence startups over the past 30 days. That's more than double what the entire U.S. venture market raised in all of 2023.

For perspective: this single month of AI funding exceeds the lifetime capital deployed to fintech, healthtech, and enterprise software combined. The numbers aren't metaphorical. This is what a capital stampede looks like.

The Numbers Behind AI's Funding Dominance

In the 30-day window from April 29 to May 29, AI and machine learning startups captured 82% of all venture funding deals. Not 40%. Not 60%. Eighty-two percent.

Out of 678 total VC deals, 556 were classified as AI-related. The remaining 122 deals—in healthtech, traditional SaaS, fintech, climate tech, and everything else—split the leftover capital.

To translate that into pure numbers: Healthtech startups closed 13 deals. Fintech managed 1. Cybersecurity got 1. SaaS captured 3. In a month when AI companies were closing deals at a rate of roughly 19 per day, every other sector was rounding to statistical noise.

The total capital deployed to AI this month: approximately $2.48 trillion. That breaks down to roughly $3.66 billion per AI deal on average. For comparison, the median Series A in traditional software is $8–$15 million. The median seed round is $1–$3 million.

The AI market is operating at a completely different scale.

Why AI Funding Is Outrunning Every Other Asset Class

Five years ago, venture capital was genuinely diversified. Healthcare got 15–20% of annual deal flow. Enterprise SaaS grabbed another 12–18%. Fintech commanded meaningful share. Consumer internet mattered. The top 10 sectors split the capital pool.

That market structure is functionally extinct. What changed? Three converging factors:

First, the technical barrier to entry collapsed. In 2020, building a large language model required a team of 50+ PhDs, custom silicon, and $100M+ in infrastructure. Today, a Stanford computer science graduate, a product designer, and a finance person can prototype a frontier AI application on cloud GPUs for $10K–$50K per month. That reduction in capital intensity from "$500M minimum" to "$20M sufficient" opened the floodgates to 10x more startups.

Second, the outcome uncertainty shifted. Five years ago, AI applications were theoretical. "We'll do NLP but better." Lots of papers, few customers. Today, the proof-of-concept is incontestable. ChatGPT serves 200 million monthly users. Anthropic and OpenAI are valued in the hundreds of billions. Anduril is selling autonomous defense systems to the U.S. military. Scale AI raised $1 billion on the strength of a code-completion tool. The upside isn't speculative—it's proven, measurable, and generating real revenue.

Third, capital got desperate. Traditional venture categories have flattened. SaaS growth decelerated. Fintech hit regulatory headwinds. Real estate tech stalled. The mega-funds—Sequoia, Andreessen Horowitz, Benchmark, Khosla Ventures—needed outsized returns to justify their fund sizes. AI is the only space where a $10M seed check could theoretically become a $50B company in three years. Everything else looks like slow money by comparison.

Combine those three factors and you get what we're seeing: not organic market evolution, but systematic capital reallocation toward a single dominant thesis. When 82% of venture money chases one category, it's not because that category is inherently 82% more valuable than everything else. It's because every other option looks worse by comparison.

Geography: The U.S. Maintains Its Stranglehold

The United States leads by a margin that's nearly impossible to close. Of 678 VC deals in the 30-day window, 282 were U.S.-based—41% of all deals. Add in the fact that U.S. deals tend to be larger on average, and the U.S. share of total capital deployed is probably closer to 55–60%.

The United Kingdom follows distantly with 69 deals (10%). India is third with 51 deals (7.5%). China and South Korea compete for fourth place with 31 and 27 deals respectively.

What's striking is that the U.S. gap has widened, not narrowed. In 2019–2020, when the AI boom started, China was positioning itself as a credible second-mover. Startups like SenseTime and others were raising massive rounds. Today, Chinese AI companies face a structural handicap: U.S. export controls on advanced chips. Without access to the latest Nvidia H100s and similar processors, Chinese AI companies can't train frontier models that match OpenAI or Anthropic. They can build applications on top of Western models, but they can't own the foundation layer.

The U.K.'s position reflects London's ascent as a credible secondary hub. Firms like Exscientia and Synthesia have raised major rounds. But even the U.K.'s 69 deals are less than 25% of the U.S. total.

India's 51 deals reflect a different dynamic: quantity over scale. Most Indian AI startups are building application-layer companies—logistics optimization, fintech rails, e-commerce recommendations. Average deal sizes are smaller ($2M–$5M) but the velocity is high. There's a long tail of 100+ early-stage Indian AI companies at the seed stage, each raising sub-$1M checks from local and international VCs.

The geographic concentration has real implications. It means talent gravity flows toward the U.S. The best ML researchers have incentives to relocate to Silicon Valley or Boston. It means corporate relationships—partnerships with Google, Meta, Microsoft—are easier to negotiate in the U.S. It means capital rounds are more competitive, better priced, and larger. Everything else chases from behind.

Funding Stages: A Market Heavy on Early Bets

The funding stage distribution reveals something important about market maturity and risk tolerance.

Of 678 AI deals: 378 (56%) were early-stage or unclassified rounds. These are typically pre-seed or seed companies—teams with an idea, a prototype, and maybe a few thousand users, but no meaningful revenue. Capital is being deployed before any real proof-of-concept exists.

Series A rounds accounted for 116 deals (17%). These are companies with some traction—proven product-market fit in a narrow segment, first paying customers, early revenue signals. Series A is where companies traditionally raise their biggest check to scale.

Seed-stage rounds: 97 deals (14%). Formal seed rounds where companies raise $2M–$10M to validate a hypothesis and extend runway.

Series B: 59 deals (9%). Companies that proved the model works and are scaling from $1M–$10M ARR to $10M–$50M ARR.

Series C and beyond: 28 deals (4%). Companies with proven, meaningful revenue, scaling internationally or preparing for exit.

The shape of this distribution is unusual. Normally, in a healthy venture market, capital flows increasingly from earlier to later stages as companies mature. Here, 56% of capital is chasing early-stage companies with unproven business models. It's not just investing in a lot of early companies—it's concentrating capital density at the earliest possible stage.

That implies two things. First, capital is being spread very thin. VCs are making more bets at smaller sizes, hedging the risk that most will fail. Second, few AI companies have actually crossed from "impressive demo" to "sustainable business" yet. Most of the Series C+ deals are winners from the 2022–2023 cohort that have held their valuations or grown. The 2024–2025 cohorts haven't had enough time to reach scale yet.

This creates a specific risk: if the AI market cools even moderately in the next 12 months, the flood of early-stage companies won't have enough runway to survive long enough to prove viability. Venture capital will rotate into "show me profitable unit economics" mode. You'd see a wave of acquihires, shut downs, and fire sales starting in Q3 2026.

Concentration Risk: A Few Winners, Thousands of Scrambling Startups

Look at the largest deals in the dataset. Anthropic's reported $65 billion round at a $965 billion valuation dominated the month. Anduril's $5 billion round at a $61 billion valuation. Cognition AI's $1 billion Series B at a $25 billion pre-money valuation (for a code-completion tool).

Three companies captured an outsized share of media attention and capital allocation. Moonshot AI's reported $2 billion round follows. Then the tail drops off sharply.

Below that you find the long tail: thousands of Series A deals in the $10M–$50M range, thousands of seed deals in the $1M–$5M range, and tens of thousands of pre-seed deals below $500K.

The median AI Series A is probably $8M–$20M. The median seed is $1M–$3M. The median "other round" is likely sub-$1M. Thousands of companies are splitting small checks while the top 0.1% absorb massive allocations.

This creates a classic problem: survivorship bias. The public narrative—"AI startups are raising $100M+ rounds all the time"—is driven by the 2–3% of companies actually raising megadeals. The other 97% are grinding on sub-$20M checks, racing to profitability before capital dries up, competing for hiring in an incredibly tight market.

What This Means for Q2 and Beyond

The AI venture boom is genuine. The capital is real. The companies have traction. But the market is also fragile.

As long as the frontier companies keep delivering new breakthroughs—new models, new capabilities, new revenue streams—capital will keep flowing into the ecosystem. VCs will justify the checks as bets on scale: "Yes, 90% of AI startups will fail, but the 10% winners will be worth it."

If momentum stalls—if there's no major model release for six months, if Anthropic hits a fundamental scaling ceiling, if U.S. export controls tighten further—capital will evaporate. You'd see Series A funding slow, seed checks shrink, and the market reprice. Some of today's "unicorns" will be worth 50% of their current valuations within 18 months.

That's not prophecy. It's pattern matching. This exact cycle happened in cleantech (2008–2011), in cryptocurrencies (2017–2018), in consumer apps (2020–2022). Massive capital concentration in one narrative, followed by a correction when the narrative cracks.

The data from this 30-day window is accurate and real. AI funding is dominating venture capital. The concentration is real. The question for the next two quarters isn't whether AI will remain hot—it will. The question is whether this specific wave of early-stage capital deployment represents sustainable growth or a front-loaded bubble that's pushing three years of investment into six months.

Based on the distribution of capital toward early-stage companies with unproven business models, the answer is probably the latter.