The agility premium: finding venture returns in the shadow of AI giants

Hello friends, and welcome to The Unsophisticated Investor! Brought to you by Mark, Head of UK Expansion & Operations at Shuttle!

If you want to invest alongside the VC funds who've backed breakout companies like Revolut, Monzo, Asana, JustEat, Bolt, Lets Get Checked, Loom, Runna, Charlotte Tilbury, Deel, Aircall, AngelList, Carta, TransferWise and many more, regardless of your knowledge, network or net worth, then sign-up today using the link below.

Now, let’s get into it 👇

In the past 18 months, a small number of AI companies have raised sums of money that now dwarf the budgets of small nations. OpenAI took $40 billion in a single round in March 2025 and then went back and raised $110 billion more early this year. Anthropic did the same. After raising $13 billion in September 2025, it took in a further $65 billion in May at a valuation approaching $1 trillion, in what is likely its last private round before a touted IPO. These are not investments in the ordinary sense.

The result is that private markets are producing an odd, counterintuitive result. The rush of money into a handful of giant AI deals is, at the very same time, the best opening the small end of the venture world has had in years.

The capital concentration at the frontier  

Start with where the US dollars are going. In 2025, close to two-thirds of every venture dollar invested in the US went into AI companies. In the first half of 2026 that figure reached somewhere around 86%. 

In the first quarter of 2026, the five largest deals in the country accounted for 73% of all the venture money invested, and across the first half of the year, rounds of $100 million or more took 87.5% of the capital, leaving everything smaller to share the remaining eighth. Venture has always been a game of a few big winners, but it has never before been a game of quite so few, quite so big.

Structural shifts in fund sizes

A company raising ten or twenty billion dollars cannot be backed by a $30 million fund. It can only be backed by a fund investing in the hundreds of millions, which means the enormous rounds can only be fed by enormous funds. 

Andreessen Horowitz (a16z) gathered something like $15 billion across its 2025 funds, by its own reckoning close to a fifth of all the venture money raised in America that year. It reportedly went looking for a further $20 billion for a dedicated AI vehicle. In the first half of 2026, just three firms, a16z, Thrive and Founders Fund, took nearly half of all the money raised by American venture funds between them. Funds of a billion dollars or more, which took about a third of all capital in 2025, took roughly 70% this year. Established firms captured over 90% of everything raised in the first quarter, up from 73.7% the year before.

The clearest picture of the distortion is in the size of the funds themselves. The average American venture fund raised in early 2026 was $289 million. The typical one, the median, was $15.3 million, down from $25 million a year earlier. 

Endowments and pension funds are among those backing this new wave of venture funding for established funds and looking to be among the backers of those AI firms with the best chance of reaching the IPO window. Pitchbook describes the concentration as structural rather than a passing phase.

The door is left open to emerging managers 

The larger funds, drunk on the AI boom and under pressure to raise ever greater sums to meet the demand OpenAI and others keep forecasting, have sent all of that capital upward and into the biggest companies at the biggest valuations. It has largely vacated the opposite end of the market. 

The seed and pre-seed rounds, the odd corners, the companies that are not building frontier AI models but putting them to work, has been left to a different kind of investor. As we covered last week in Part 4 of the Evolution of Venture Capital, the market resembles a barbell where mega-funds are thriving at one end, specialist small funds thriving at the other, and the traditional mid-sized fund stuck in the middle with the stiffest headwinds of the lot. The money piling up at the top has, almost as a side effect, cleared room at the bottom.

The micro-fund advantage 

What moves into that space has an advantage the giants, for all their billions, structurally cannot match. Let’s call it the agility premium.

Some of it is pure arithmetic. It is far easier to treble a $10 million fund than a $10 billion one, because a single good outcome can return the whole of a small fund and barely registers on a large one. A small manager writing small cheques into cheap early rounds can own a meaningful slice of a company for very little and needs only one of them to come good. Some of it is speed. A solo investor can decide in an afternoon what a large partnership takes several weeks with an investment committee to approve, and the best early deals do not typically sit around that long. 

Some of it is focus. A manager who does nothing but developer tools, or nothing but robotics, tends to meet the best founders in that niche before anyone at a generalist fund has heard the name.

The big firms keep real advantages that agility does not cancel out. Their names alone help a startup raise its next round, and for some founders, the appeal of the brand and their PR machine can be too good to pass on. They also have the deep pockets to keep funding their winners round after round, defending their stake while a small fund is quietly diluted away. They can lead the giant late-stage rounds a small fund can only watch. And they employ whole teams to help founders hire, sell and market, which a solo GP cannot. Agility wins the first cheque, but it does not guarantee that you win the next ten.

Historical drivers of venture returns

Over time, the small end has earned its keep. Cambridge Associates, which has spent years taking apart where venture returns actually originate, found that over the past decade between 40 and 70% of the industry's total gains were produced by new and emerging managers rather than the established houses. Since 2005, funds smaller than $500 million have accounted for at least half of the gains in venture's hundred best investments, and in five of those years more than 60%.

This is not a new idea. The Kauffman Foundation, in its famous autopsy of its own venture portfolio, concluded years ago that the smaller funds were the ones actually earning their fees, while too many of the big-name funds struggled to beat what you could have made in the public markets. What is new is that the AI boom has widened the door to the part of the market where that pattern has most reliably held.

Meanwhile, in Europe

You can watch the agility premium being collected right now in Europe, where a handful of solo investors are backing the early-stage companies the giants arrive too late for.

Nathan Benaich, who writes the annual State of AI report, closed his third fund of his Air Street Capital at $232 million, the largest solo GP fund ever raised on the continent. His early bets include Synthesia, the video company now doing more than $150 million a year; Black Forest Labs, whose image models sit behind a great deal of content on social media and Poolside, a frontier lab. He was in early in each of these, when the cheques were still small.

Carles Reina's Baobab Ventures is smaller, at $15 million. But Reina wrote the first cheque into ElevenLabs, the voice-AI company that became one of Europe's fastest risers, well before that was obvious. He has backed more than seventy companies and via Baobab invests into AI, robotics and defence, the corners European institutions still shy away from.

There’s another crop of emerging managers and micro-funds / solo-GPs coming out of Ireland too. Finn Murphy’s Nebular that he runs out of London and New York, was among the first investors into Starcloud, the company building AI data centres in space, and recently valued at $2.3 billion and founded just two years ago. 

Others include Declan Kelly, an early Web Summit hand, who built his Foreword fund on a network of more than 80 founders and operators, which got him in early on Sorare, Wayflyer and Localyze. And Baseline Ventures led by Eamon Leonard and Fiona Kelly, Ireland's only dedicated and privately backed pre-seed fund, investing €100,000 as first cheque investors into pre-seed companies.

These investors are among a new generation of venture funds that are able to leverage sectoral expertise, and as ex-founders themselves, the practical know-how of how to help support the next generation of founders. 

The outlook for early-stage capital

The unprecedented concentration of capital at the top of the market is ultimately a story of institutional anxiety rather than guaranteed returns. While this pile-up of capital has made life difficult for those raising mid-sized conventional funds, it obscures a reality that capital density does not dictate where the next generation of outlier companies will originate. The agility premium certainly comes with varying degrees of success, but it provides a structural advantage that incumbents can’t easily replicate. 

By chasing the AI giants, mega-funds have inadvertently left the ground floor of venture capital quieter, cheaper, and more open than it has been in years. If historical cycles of capital allocation hold true, this market distortion offers an opportunity for micro-funds willing to step into the gap. Away from the noise of multi-billion-dollar valuations, the foundation for the next decade of venture returns is being quietly assembled, one small cheque at a time

What we’ve been working on at Shuttle

• Putting the final touches on upgrades to our platform 🧑‍💻

• Planning our UK launch events & comms 🏙️

The Unsophisticated Investor is brought to you by Mark, Head of UK Expansion & Operations at Shuttle. Shuttle's mission is to break down the barriers to private markets and make wealth-building opportunities accessible beyond the ultra-wealthy elite. To do it, we're building Europe's private market infrastructure platform for running investment operations. By stripping away the legal friction and admin for deal managers, we make it easier than ever for high-performing opportunities to open up to a wider, modern network of investors.

Mark
Shuttle’s Head of UK Expansion & Operations