Global startups pulled in a record $510 billion in the first half of 2026 — more than investors deployed across all of 2025 — as a handful of frontier artificial intelligence labs bent the entire venture market around their fundraising, according to Crunchbase data released July 2.
The figure shatters every prior half-year mark, eclipsing the $375 billion peak set at the height of the 2021 boom. But the headline number obscures how narrow the surge has become. OpenAI and Anthropic alone accounted for $217 billion — 43% of all startup funding in H1 — a level of concentration in two companies that has no precedent in the history of venture capital.
"The defining characteristic of venture investment in the AI boom remains its extraordinary concentration in terms of companies, industries and geography," wrote Gené Teare, the Crunchbase senior data editor who compiled the report. Close to a third of all Q2 global venture dollars went to a single company: Anthropic, which raised $65 billion in the quarter and overtook OpenAI to become the most valuable private company on The Crunchbase Unicorn Board.
The numbers behind the record
Investors poured $305 billion into more than 5,000 startups in the first quarter and another $205 billion in the second, making Q1 the largest quarter on record and Q2 the second-largest. More than 70% of global startup capital in Q2 flowed to AI-focused companies, up from just under 50% a year earlier. In the United States, which absorbed two-thirds of global funding in the quarter, AI's share of venture dollars ran even hotter — around 86%, according to reporting on the U.S. breakdown.
The mega-round has become the market's organizing unit. In Q1, four frontier labs — OpenAI ($122 billion), Anthropic, xAI ($20 billion) and self-driving company Waymo ($16 billion) — collectively raised roughly $188 billion, or about 65% of all global venture investment in the quarter. In Q2, 16 companies raised billion-dollar rounds totaling $108.6 billion, more than half of all funding for the period. Seven of those were frontier AI labs, including China's DeepSeek, StepFun and Moonshot AI, the U.K.'s Ineffable Intelligence, and U.S.-based Prometheus and Isomorphic Labs.
Crucially, the boom is also spilling outward. Billion-dollar rounds landed in defense, AI infrastructure, robotics and healthcare — a sign, Crunchbase noted, that the investment wave "has grown well beyond a select few top foundation labs." Late-stage funding hit $134 billion in Q2, up 141% year over year, and the number of companies raising $100 million-plus Series A and B rounds climbed to 91 globally.
Liquidity returns
For the first time since 2021, the exit door swung open alongside the funding spigot. Q2 produced the strongest venture-backed exit market in years, anchored by two records: SpaceX went public at a $1.77 trillion valuation, raising $75 billion in the largest IPO ever for a venture-backed company, then moved within a week to acquire Anysphere — maker of the AI coding tool Cursor — for $60 billion, the largest startup acquisition on record.
In all, 32 companies went public above $1 billion in Q2, including inference chipmaker Cerebras Systems and quantum firm Quantinuum, while 24 companies were acquired at or above $1 billion for a combined $113 billion — the highest quarterly acquisition total Crunchbase has tracked. The reopening of public markets matters because it gives investors a way to recycle capital rather than simply pouring more in at the top.
Why this matters
A record built on two companies is a fragile kind of record. The same concentration that produced the $510 billion headline is what worries the people who allocate the money. Strip out the four largest mega-rounds — OpenAI, Anthropic, xAI and Waymo — and comparable venture activity tracked close to ordinary 2024 and 2025 levels. In other words, most of the "boom" is two logos.
That creates a systemic exposure problem that individual deals hide. Because general partners underwrite each transaction on its own merits, limited partners can end up quietly stacked into the same names across multiple funds. "They don't think about the underlying exposure that an LP might be in all three of those firms," Darius Craton, a director at Raymond James Private Capital Advisory, told analysts examining LP risk — warning that backers should weigh the concentration they may be accumulating without realizing it.
The capital is also starving the rest of the ecosystem. Institutional LPs directed 91% of new first-quarter commitments to brand-name firms, up from 74% a year earlier, while emerging managers saw fundraising fall 35% to roughly $12 billion — a dynamic investors like Rick Zullo of Equal Ventures have flagged as a structural threat to venture returns five to seven years out. At a StrictlyVC panel in Athens in May, senior investors from Verdict Capital, Threshold Ventures and Atomico noted that roughly 75% of all 2026 venture capital was flowing to just five AI companies, calling the pattern unprecedented.
Whether this is a durable repricing of the economy or the biggest bubble in venture history is the open question. The bull case is that AI genuinely represents, as one analyst put it, "the largest wealth creation event in human history." The bear case is that valuations set by a handful of strategic backers — Nvidia, Microsoft and Amazon increasingly anchor these rounds as capital-markets events rather than traditional venture syndicates — are circular, propping up demand for the very infrastructure the labs are buying.
What to watch
The next test is whether the exit market holds. If IPOs and M&A keep clearing at 2026's pace, record private investment and real liquidity could reinforce each other into a sustainable cycle. If the exit window narrows while mega-rounds keep inflating private marks, the gap between paper valuations and realized returns becomes the story. Watch three things through the back half of the year: whether AI's share of dollars stays above 70%, whether billion-dollar rounds keep broadening into defense, robotics and healthcare, and whether any frontier lab's valuation stalls. In a market this concentrated, the second half of 2026 may hinge less on how much startups can raise than on whether the biggest names can grow into the numbers already written on them.
"The defining characteristic of venture investment in the AI boom remains its extraordinary concentration in terms of companies, industries and geography."-- Gene Teare, Senior Data Editor, Crunchbase News