The numbers: what Crunchbase actually recorded
Crunchbase published quarterly and half-yearly data in 2026 that set historical records across multiple dimensions. Q1 2026 was the largest venture capital quarter ever recorded: US$ 300 billion invested in approximately 6,000 startups globally, growth of over 150% both relative to the previous quarter and to the same period in 2025. In a 90-day window, global venture deployed approximately 70% of all 2025 capital.
H1 2026 (first half) consolidated this level: US$ 510 billion in total, with Q2 being the second-largest quarter in history (US$ 205 billion across more than 5,000 startups), after Q1. H1 2026 surpassed the previous peak of H2 2021 (US$ 375 billion) — the peak of the post-pandemic startup boom.
But the nature of this growth is radically different from 2021. Silicon Canals (July 2026) captures the distinction: "Q1 2026 was the largest and most concentrated venture quarter ever recorded — approximately US$ 300 billion deployed, about US$ 242 billion of it to AI, and close to US$ 188 billion of that to just four companies. The right way to read this is not as a broad recovery, but as a barbell."
The concentration: four companies, 65% of the capital
The most revealing data point from Q1 2026: OpenAI (US$ 122 billion), Anthropic (US$ 30 billion), xAI (US$ 20 billion), and Waymo (US$ 16 billion) collectively raised US$ 188 billion — equivalent to 65% of all global venture capital in the quarter. Four rounds among the five largest venture rounds ever recorded were closed in Q1 2026.
To provide context: in one quarter, just two frontier AI lab companies (OpenAI and Anthropic) raised US$ 152 billion — more than the entire global VC market raised in any quarter prior to 2021. OpenAI and Anthropic together represented 43% of all startup funding in H1 2026 (US$ 217 billion out of US$ 510 billion total, according to Crunchbase).
AI Insider (May 2026) describes what is happening accurately: "The days of funding AI wrappers — products that are essentially thin interfaces on top of someone else's model — are largely over. What is left is a barbell: a massive, capital-intensive frontier on one end, a regular-sized venture market on the other, and very little in the middle."
What is outside the megarounds
The concentration at the top does not mean the broader ecosystem is collapsing. The analysis by Qubit Capital (July 2026) shows that fundamentals still exist: approximately US$ 72 billion were deployed across about 4,595 deals outside the megarounds in Q1. Early-stage grew 41% year-over-year — more moderate than late-stage, but still solid by historical standards.
Kleiner Perkins launched a US$ 3.5 billion fund dedicated exclusively to AI startups — one of the largest AI-focused vehicles ever raised by a single firm. Abu Dhabi's MGX fund closed at US$ 49 billion, above its US$ 45 billion target. Tech defense capital reached US$ 12.3 billion in H1 2026 — almost double the previous year's total.
For founders outside the frontier labs, AI Funding Me (May 2026) identifies the sectors attracting capital with rising valuations: AI infrastructure (data centers, compute, networking), vertical AI for specific industries (healthcare, legal, finance), AI for defense and security, and robotics.
What investors are really pricing in
The concentration in OpenAI, Anthropic, xAI, and Waymo reveals what the largest-scale investors are buying: not just products, but positions in infrastructure that others will build upon. The reasoning is similar to what justified investments in AWS and Google Cloud: whoever controls the compute layer and the foundational models controls the platform upon which the next generation of software will be built.
Abu Dhabi's MGX is developing Europe's largest AI campus near Paris with 3GW of compute capacity — treating AI infrastructure as a sovereign wealth fund asset, not traditional venture. AI Funding Me documents: "The sovereign fund market treats frontier AI infrastructure as a sovereign asset class, not traditional venture capital."
For the rest of the ecosystem, the practical implication from Deloitte (cited by Index.dev): 74% of the economic value generated by AI is captured by 20% of companies. The US$ 100 billion rounds are not the market — they are the distortion that defines the context in which the real market operates.
