AI concentration fractures venture capital, leaving smaller funds and non-AI startups in distress

AI Concentration Is Fracturing the Venture Capital Industry

The venture capital industry is undergoing a structural rupture, with artificial intelligence investments funneling extraordinary sums into a tiny number of firms while leaving the rest of the ecosystem starved of capital and exit opportunities. The numbers make the imbalance stark: just five companies — OpenAI, Anthropic, Elon Musk's xAI, Alphabet's autonomous vehicle unit Waymo, and data center operator Nscale — absorbed 78% of all venture deal value in the first quarter of 2026, according to PitchBook data. The IPO of SpaceX, which now owns xAI, alone represented roughly 73% of all exit value in the same period. The firms positioned early enough and large enough to back these giants, chiefly Andreessen Horowitz and Founders Fund, now sit atop a winner-take-most hierarchy that most venture managers simply cannot access.

The structural problem is one of check size and brand. Leading a seed round in a competitive AI company now requires writing a $10 million to $15 million check as a minimum, according to David Clark, Chief Investment Officer of VenCap International, which has itself invested in Andreessen Horowitz alongside roughly a dozen other VC funds. For a $100 million fund, that commitment is existentially large. Smaller and emerging managers — defined as those operating three or fewer funds — raised about $62 billion last year, roughly 60% below the $163.4 billion they captured during the 2022 pandemic peak. Established managers fared better in absolute terms but still raised only about a third of their 2022 totals, pulling in $84 billion last year.

Performance divergence is accelerating alongside the fundraising gap. Carta data shows that the spread between top and bottom fund performers has more than doubled for 2026 vintage funds compared with those formed between 2017 and 2021. The top decile of 2024 vintage funds posted gains of 49.3% in the fourth quarter of last year; the bottom quartile lost 17.4% in the same window. Limited partners — the institutional investors, endowments, and family offices that provide VC capital — are responding rationally if ruthlessly, demanding proof of returns and liquidity before committing to new funds. Felix Capital, the London-based firm known for backing Peloton and Deliveroo, set out to raise $600 million for its next fund and remains $150 million short. German firm 468 Capital abandoned plans for a $1 billion growth fund entirely and will pivot to a smaller early-stage vehicle. Northzone, a 30-year European institution managing $3 billion in assets, is assembling its newest fund from a broader-than-usual collection of investors writing smaller checks.

The Zombiecorn Problem and Fintech's Painful Reckoning

The AI concentration is creating collateral damage far beyond struggling fund managers. A generation of startups that reached billion-dollar valuations during the 2020–2022 boom now occupy an uncomfortable limbo the industry has labeled "zombiecorns" — companies whose growth trajectories no longer justify their peak valuations and that lack viable paths to IPO or acquisition. The problem is especially acute in enterprise software, where investors fear that AI tooling from OpenAI, Google, and Anthropic could render entire product categories obsolete. Personio, a German HR software company that was valued at $8.5 billion in 2022, is reportedly preparing to raise new capital at roughly half that figure.

Fintech, once the most celebrated sector outside pure consumer tech, has bifurcated sharply between a small group of scaled survivors and a large cohort of struggling mid-tier players. Global funding for fintech startups exceeded $100 billion in 2021; that era now looks like a historical anomaly. The companies thriving today are those that achieved genuine scale. Stripe processed $1.9 trillion in payments in 2025, up 34% year over year, and is valued at $159 billion. Revolut, profitable and Europe's most valuable private startup at $115 billion, is planning a public listing within a few years. These outliers, however, are not representative of the sector. Monzo, a UK digital bank that is profitable but valued at a fraction of Revolut's figure, explored a secondary share sale last year and abandoned the effort. Atom Bank has been unable to attract viable buyout bids.

Institutional investors in fintech are absorbing visible losses. Chrysalis Investments, a listed UK venture fund, announced it was winding down in February after years of underperformance that required writedowns on positions including UK digital bank Starling Bank and insurance technology firm Wefox. Augmentum, another UK-listed fund that backed fintech unicorns including digital lender Zopa Bank, was taken private at a discount of roughly 30% to net asset value. Augmentum's chief executive Tim Levene acknowledged on an earnings call that selecting only winners is an impossible standard in venture investing — a candid concession that the sector is in a reset.

What Comes Next for Funds and Founders

The path forward for both fund managers and founders in non-AI sectors is narrowing. Acquirers are growing more selective, IPO windows remain difficult, and the alternative — waiting indefinitely — depletes investor patience and employee morale. Gulsah Wilke, a partner at DN Capital, has cautioned that founders and investors may need to recalibrate expectations entirely, accepting that transformative exits may never arrive for many companies that once seemed destined for them.

There are pockets of adaptation. Tapestry VC, a smaller fund, parlayed an early investment in Fin — an AI-powered customer service company acquired by Salesforce for $3.6 billion — into the credibility needed to close its third fund at $80 million. The deal demonstrates that smaller managers can still generate outsized returns, but doing so increasingly requires either early AI exposure or exceptional timing in secondary sales and acquisitions. Kyle Stanford, director of US venture capital research at PitchBook, captured the new calculus bluntly: the 2021 narrative around neobanks serving the underbanked simply cannot command the growth multiples that investors now associate with companies like Databricks or Stripe. For the many startups and funds still operating on 2021-era assumptions, the adjustment is likely to be prolonged and painful.