Founder-Investor Power Balance Shifts as Startup Funding Hits $392B

North American startups raised $392 billion in the first half of 2026, according to Crunchbase data published in early July. That figure, reported by Joanna Glasner, shatters every prior six-month record and nearly matches the full-year totals of the 2021 boom, the last period anyone described as a founder's market. If the raw numbers told the whole story, founders would be walking into every term sheet negotiation with the upper hand. They do not tell the whole story.

The quarterly data now shows a market bifurcating along lines that most round announcements do not capture. A tiny cohort of AI companies absorbed hundreds of billions in capital, Forbes reported in May, rewiring the distribution of venture dollars. At the same time, early-stage founders in Canada raised 40 percent less capital in the first quarter of 2026 than they did a year earlier, according to RBCx data released in June, even as their funding needs held steady. The balance of power is not tilting in one direction. It is fracturing along fault lines that separate the sought-after from everyone else.

On the investor side, the concentration of capital has accelerated to a degree that even industry veterans find striking. Inc. reported in late June that six venture capital firms now dominate startup financing at a scale that reorders the negotiating table for founders. Umesh Padval, a two-decade veteran of the industry, told the magazine that the current landscape has no precedent.

I've never seen it before over 20 years in the industry., Umesh Padval, venture capital investor, as quoted by Inc.

What Padval was describing is the rise of the giga-VC, a class of firm that can write a $100 million check without assembling a syndicate and, critically, can do so on its own timeline. When six firms control the bulk of deployable capital, founders who need to raise are negotiating with a narrower set of counterparties. That narrows the competitive dynamic that historically gave founders leverage on price and terms. A founder raising a Series B in enterprise SaaS is not playing six firms off each other. She is playing two or three, and they know it.

The megafund phenomenon has downstream effects that reach limited partners as well. CNBC reported in mid-July that gaining access to the large funds, and paying their steep fees, has become difficult even for sizable family offices. When capital concentrates at the top of the GP stack, the LP base narrows with it. The result is an ecosystem where both the supply of startup capital and the governance of that capital are managed by an increasingly small number of institutions, each with its own internal politics and portfolio-construction logic.

The governance question has broken into public view through a different channel: the IPO market. When SpaceX filed to go public, the company disclosed a dual-class share structure granting Elon Musk outsized voting control, Reuters reported in May. The filing revived a debate that had been dormant since the 2023 downturn reset expectations. Dual-class structures are not new, but SpaceX's deployment of the mechanism at its scale, and in the current regulatory climate, signals something about founder leverage at the top of the market. The strongest founders are not merely raising on favorable terms. They are building permanent governance firewalls.

The New York Times reported in late June that SpaceX had joined a growing cohort of companies using dual-class share structures to ensure many shareholders have little say in how the business is run. From the founder's perspective, the logic is straightforward: if the public markets will punish long-term capital allocation with quarterly myopia, the rational response is to structure the company so the founder can ignore the shareholders. The question the venture industry is now debating privately is whether that logic migrates backward into the private markets, where a generation of founders who watched the 2021-era class get diluted into irrelevance is now demanding similar protections at the Series C and D stages.

The Khosla Ventures approach to Runlayer's Series A, detailed by Fortune in late June, is the mirror image of the same dynamic. According to Fortune's reporting, Vinod Khosla wanted "every available dollar" of the round, a move that would have concentrated the cap table under a single large investor rather than dispersing it across a traditional syndicate. Runlayer, an enterprise AI startup, ultimately closed a $30 million Series A co-led by Felicis and Khosla Ventures, according to a June announcement. But the negotiation itself revealed the tension at the heart of the current market: the investors with the deepest pockets want allocation concentration, not diversification, and they are willing to pay up for it.

For founders, the Khosla maneuver cuts both ways. A single large investor can simplify board dynamics and accelerate decisions in a way that a four-firm syndicate cannot. But it also concentrates power on the other side of the table. If that investor sours on the strategy, or if the partnership dynamics at the fund shift, the founder has no second opinion to appeal to, no competing voice in the boardroom. The safe harbor of a diversified cap table, long treated as standard governance hygiene, is now a luxury some founders are trading away in exchange for speed and valuation.

The geographic concentration of capital reinforces the asymmetry. The Los Angeles Times reported in July that California drew ten times more venture capital than any other state in the first half of 2026. Founders outside the Bay Area, and outside a handful of AI sub-sectors, are not operating in the same market as the headline numbers describe. The $392 billion figure aggregates activity across two universes that barely touch: the AI capital-intensive frontier and the rest of the technology economy.

The quarterly shifts are measurable. Nearly 90 unicorns were minted in the first half of 2026, according to an Under30CEO analysis published in July, a pace that rivals the frothiest periods of the last cycle. But the distribution is lopsided. Forbes, in its March 2026 State of Venture Capital report, described the moment as a "value creation era" in which AI has moved from hype cycle to core infrastructure and investors are shifting from growth-at-all-costs to selective capital deployment. The phrase "value creation" sounds benign, but in the context of a VC market, it is a euphemism for terms that protect the investor when growth slows.

What that means in practice is that participation rights, liquidation preferences, and anti-dilution provisions are creeping back into term sheets after the founder-friendly era of 2020 and 2021 largely stripped them out. Lawyers who draft these documents, speaking at industry panels over the past quarter, have noted a revival of structured terms that allocate downside protection to investors in exchange for the headline valuations founders want to report. The announced number looks strong. The unannounced mechanics tell a different story.

What Changed This Quarter

The second quarter of 2026 introduced several accelerants. The SpaceX IPO filing in May normalized dual-class governance for a new cohort of late-stage founders, providing a template that lawyers can adapt for Series D and E rounds. The Runlayer negotiation demonstrated that top-tier investors are willing to compete aggressively for allocation in AI infrastructure deals, but on terms that concentrate their own influence. And the Crunchbase data confirmed that the aggregate numbers are being driven by a shrinking set of companies and geographies, even as the headline figures suggest a rising tide.

Founders Fund's $600 million bet on SpaceX, which became a windfall exceeding $50 billion when the company went public, TechTimes reported in mid-June, has reinforced the conviction among large LPs that concentration, not diversification, is the path to venture returns. That conviction flows downstream. If LPs reward concentration, GPs will concentrate. If GPs concentrate, founders face fewer options and more powerful counterparties.

The counterweight, for now, is the IPO window. The SpaceX listing, alongside a handful of other large public offerings in the second quarter, has given late-stage founders an alternative to staying private and accepting structured terms from the same set of crossover funds that dominated the last cycle. The public markets are absorbing equity at volumes that were unavailable in 2023 and 2024, and that liquidity option restores some founder leverage at the margin. Whether that window stays open through the third quarter depends on interest-rate expectations and the performance of the 2026 IPO cohort in its first two earnings cycles.

For founders raising at the seed and Series A stages, the calculus is different. The data from RBCx on the Canadian market is a leading indicator of what happens when capital concentrates geographically and thematically: early-stage dollars contract even as late-stage aggregates soar. Founders who cannot credibly position themselves as an AI infrastructure play, or who are building outside the Bay Area, are encountering a fundraising environment that feels nothing like a $392 billion market. They are raising smaller rounds, on less favorable terms, from a narrower set of funds, and they are being asked to show metrics, revenue, and capital efficiency that were optional in 2021.

The third quarter will test whether the bifurcation widens further. Earnings from the newly public class of AI companies will either validate the private-market valuations that preceded them or trigger a repricing that cascades downward through the Series D, C, and B tiers. Founders who accepted structured terms in exchange for headline valuations will learn whether those structures activate. And the six firms that now dominate startup financing will either deploy their record fund sizes or slow their pace, signaling whether the concentration trend is a cycle or a permanent reordering of the venture capital industry.