Each quarter, we want to share how we see the market, what we observe from founders, and how those observations shape the way we build StudioAlpha.

This is a market letter from a venture firm. We start with macro. Rates, liquidity, energy, geopolitics, and public-market sentiment shape what founders can raise, what customers buy, and what LPs are willing to commit.

The short version: venture capital is under pressure. Not because startups stopped mattering. They matter more than ever. But the old model — raise a fund, chase hot software categories, make a few introductions, wait ten years — is no longer working.

AI lowers the cost of building software. Distribution is harder. Product moats decay faster. Liquidity remains slow. Capital is concentrating into fewer companies, fewer managers, and fewer geographies.

So the question is not whether venture capital survives.

The question is: what kind of venture capital still works in this market at all.

Subscribe now

1) Macro matters because it sets the cost of ambition

The global operating environment has become less forgiving.

The IMF projects global growth of 3.1% in 2026 and 3.2% in 2027, while warning that downside risks dominate: geopolitical conflict, trade tensions, fragile public debt, and weaker-than-expected AI productivity gains could all pressure growth and financial markets.

The important point is not the exact growth forecast. It is the direction of pressure. Higher rates keep capital more expensive. Energy and geopolitical shocks make planning harder. Slower growth makes customers more cautious. And when public markets are nervous, LPs become slower to commit to illiquid assets like venture. Macro does not decide which founder wins. But it changes how much time, money, and trust founders get to prove they are right.

For founders, this matters in practical terms.

Higher long-term rates raise the discount rate for future cash flows. That lowers the value of long-duration assets — including venture-backed startups — and keeps investors more selective.

Customers are still buying. But they are buying with more scrutiny, too. Budgets are not gone, but approvals are harder. CFOs want clearer ROI. Products that save time, reduce cost, or create measurable revenue still get attention. Nice-to-have software sits in procurement purgatory, where many SaaS dashboards go to die quietly.

That is the chain: macro affects liquidity, liquidity affects venture, venture affects founders, and founders decide what actually gets built.1

2) Europe is under pressure

Europe’s weakness has been building for years. Jamie Dimon recently put the gap bluntly: Europe’s GDP has fallen from roughly 90% of the US level to about 65%.2 IMF work points in the same direction on living standards: Europe’s GDP per capita at purchasing-power parity is now around 72% of the US level, with lower productivity explaining most of the gap.3

The latest growth numbers fit that pattern. Eurostat reported Q1 2026 GDP growth of only 0.1% in both the euro area and the EU. Germany grew 0.3%, France was flat, and Italy grew 0.2% quarter-on-quarter.4

The deeper problem is years of underinvestment in defence, infrastructure, energy independence, growth companies, and the core AI stack: chips, compute, frontier models, and hyperscale platforms. Europe has strong technical talent and important companies such as ASML and Mistral. But it has not yet built a competitive AI industrial base comparable to the US or China. Europe’s weak growth is not one bad quarter. It is a symptom of a system that has been too slow to adapt.

Europe does not only have a capital gap. It has an operator gap. According to Plural, only 8% of European investors have worked in a fast-growing tech company. That matters. Venture is the riskiest asset class in the world, yet much of Europe’s early-stage capital is allocated by people who have never built, scaled, or fought inside the kind of company they are judging.5

Public capital has helped build the European ecosystem. The European Investment Fund is the continent’s most important LP by far and deployed more than €7bn in equity investments in 2024 through financial intermediaries. That support matters. But it also creates a risk: when public money becomes too central, venture can start looking less like founder selection and more like policy allocation.6

If this is a bubble, it may not pop loudly. It may deflate quietly: weak DPI, slow write-downs, zombie funds, polite annual reports, and billions of paper value that never turns into cash. The market will not punish everyone at once. It will simply stop rewarding generic capital. When paper value disappears, that capital cannot be recycled into the next generation of European funds and startups. The ecosystem then weakens quietly: fewer distributions, fewer new commitments, fewer serious managers, fewer fundable founders.

3) Capital Is Concentrating

The US has the deepest venture market, but it is not a free-money market.

The Federal Reserve described US economic activity as expanding at a solid pace, while inflation remained elevated. It kept the federal funds target range at 3.50% to 3.75%. That matters for venture: higher rates keep risk appetite selective, LPs cautious, and fundraising harder for average companies.7

So the venture market is not closed. But the headline numbers are misleading.

CB Insights reported record quarterly venture funding of $285.5B in Q1 2026. But 43% came from one company: OpenAI. Deal count fell to 6,598, the lowest level since Q4 2016.

The picture is clear: more capital, fewer deals, larger rounds, stronger concentration.

AI captured 79% of all venture funding in Q1. Mega-rounds of $100M+ represented 86% of all funding. In Q1, the US captured 83% of global venture funding: $236.1B out of $285.5B.

In AI, according to CB Insights’ State of Venture webinar, the gap was even sharper. The US captured about 91% of AI venture funding worldwide. Europe captured roughly $10.5B versus about $206B in the US — around 20x less.8

This matters because VC has not earned the benefit of the doubt. Over the past decade, median venture performance has looked weak versus public markets, especially after fees, illiquidity, and slow distributions. Axios, citing Cambridge Associates data, reported that median US VC underperformed broad public market indices over the 1-, 3-, 5-, and 10-year periods, and that median DPI remained below 1x for every vintage since 2015. In simple terms: a lot of venture has produced paper value, not cash back.9

That is why “more capital into startups” is not enough. The industry needs better selection, better company-building support, and better discipline.10

That is founder math.

For ambitious AI founders, ignoring the US capital market is not a lifestyle choice. It is a strategic disadvantage. You need to understand where the capital, customers, talent, and follow-on investors are concentrated.

For investors, the implication is different.

If everyone is chasing OpenAI, Anthropic, and the obvious AI winners, the real opportunity is earlier: before consensus capital forms, before valuations detach, before every investor presentation sounds the same.

Generic VC struggles in this market because access to capital alone is no longer enough. The firms that matter will need sharper selection, stronger technical judgment, better founder support, and real access to follow-on capital.

4) Customers do not buy features. They buy outcomes.

AI is reducing the cost of producing software features. That changes the signal investors and customers should read from a product.

We see this directly in our portfolio. The strongest founders are no longer trying to impress customers with AI features. They are trying to prove a measurable outcome: more revenue, lower cost, less risk, faster execution.

In the previous software cycle, a working product carried more information. It suggested technical competence, execution speed, and some degree of product judgment. In the AI cycle, that signal is weaker. More teams can produce similar functionality faster.

The scarce asset moves elsewhere.

The relevant question is no longer whether a product can generate a document, answer a question, summarise a file, or automate a task. Those are outputs. Outputs are useful, but they are not the same as business value.

The harder question is whether the output changes the economics of the customer.

A legal AI product is not interesting because it drafts a memo. It becomes interesting if a law firm can handle more matters with the same team, faster turnaround, and fewer drafting errors.

A sales AI product is not interesting because it writes replies. It becomes interesting if it increases conversion, reduces support cost, or shortens the sales cycle.

That distinction matters.

• A feature is capability.

• An output is produced work.

• An outcome is economic change.

The best AI-native business software will not be judged by how clever the interface feels. It will be judged by whether it becomes part of the customer’s operating system: lowering cost, increasing revenue, reducing risk, or improving speed in a way the customer can measure.

The best systems will also measure this continuously. They will not only do the work. They will show the customer what changed: hours saved, cost reduced, conversion improved, revenue added, risk lowered. In other words, the software starts to prove its own value.

The moat also changes. In the old software cycle, the product itself often carried more defensibility because building it was hard. In the AI cycle, the first version is easier to copy. Defensibility has to come from somewhere else: proprietary workflow data, customer trust, integration depth, distribution, domain expertise, and becoming part of how the customer operates.

With AI a feature can be copied more easily. A workflow position is harder to copy. A measured outcome that improves over time is harder again.

That is the filter.

B O X

How to Invest in AI

For a deeper view on how we think about AI investing across chips, infrastructure, models, software, agents, and company-building, see our Pulse essay: How to Invest in AI — A field guide for a moving target

5) What our portfolio is teaching us

A few lessons are becoming clearer.

1. First, AI does not remove the need for founder quality. It increases it. More people can build a product now. Fewer can build a company.

2. Second, speed matters, but direction matters more. The best founders do not just move fast. They learn fast. They change their mind when the market gives them better information.

3. Third, early revenue is useful, but not all revenue is equal. A customer who experiments with a tool is not the same as a customer who depends on it. We care about signs of dependency.

4. Fourth, technical depth is becoming more important again. The first wave of AI products could impress with interface and novelty. The next wave will need architecture, data strategy, integration, reliability, and domain-specific performance.

5. Fifth, fundraising is now a proof game. The old version was: tell a big story, show a large market, raise the next round. The new version is stricter: show why this team, why now, why this workflow, why customers care, and why the company can become difficult to copy.

That rewards operator judgment.

6) What we are building at StudioAlpha

StudioAlpha is built around four functions: access, judgment, production, and capital intelligence.

Access means seeing strong founders early. We source through operator networks, founder relationships, and our content engine. StudioAlpha’s publication reaches more than 105,000 subscribers and 1.9 million views and gives us direct visibility into founders, operators, and investors who already think about AI, software, and company-building. Combined with our San Francisco network and European operator base, this creates deal flow before many companies are formally formed or fundraising.

Judgment means applying a stricter filter before the market does. We look for founder quality, technical depth, workflow relevance, measurable customer outcomes, and signs that the company can become difficult to copy.

We also use our own AI-augmented selection infrastructure. Every founder team entering our funnel is assessed through a structured process that combines application data, diligence inputs, founder interviews, market research, and partner scoring into one internal analytical layer.

AI helps us stress-test market claims, compare companies against their competitive landscape, capture interview insights, and surface patterns that a purely manual process may miss.

It does not replace investor judgment. It makes our judgment more systematic.

Production means working with founders after investment. The same internal knowledge infrastructure supports our Production Program: sharper positioning, better customer evidence, stronger technical decisions, cleaner fundraising materials, and preparation for US investor standards.

Capital intelligence means understanding where AI value is moving across public markets, private markets, infrastructure, software, and founder formation. A day-zero investor cannot only read startup decks. The surrounding capital system matters. This is why we started building the portfolio with our own money before raising the fund.

B O X

Portfolio snapshot

We built the first StudioAlpha portfolio before inviting LPs into Fund I. The point was to test the strategy with our own capital first.

Current portfolio

• 15 companies

• Focus: AI-native business software and workflow-heavy verticals

• Key verticals: legal AI, wealth technology, recruiting, e-commerce AI, production/media, sales, staffing, hospitality

Founder origin

• USA: ~50%

• Europe: ~30%

• Latin America / Asia: ~20%

Commercial traction

• At initial investment, 4 of 15 companies had revenue.

• Today, all 15 have revenue or active commercial traction.

The pattern is still early, but clear: the strongest companies are moving from AI demos to measurable business outcomes. We will report portfolio development more explicitly in future quarterly letters.

Contact us

For LPs, pre-seed investing has historically been the most artisanal stage of venture capital: high-volume, low-information, and heavily reliant on individual partner intuition. We are building StudioAlpha around a different premise: disciplined data capture, combined with generative AI, can make pre-seed investing more repeatable, more systematic, and more compounding.

From the first founder application to the final push toward seed-round readiness, the same proprietary AI infrastructure supports each stage of our process. It helps us evaluate more opportunities with greater rigor, support portfolio companies with faster feedback, and learn systematically from every company we assess and every investment we make.

For LPs, this creates three concrete advantages:

• better selection through our AI-augmented selection process

• stronger operational support after investment

• a fund edge that compounds with every deal we do

That is the model we are building.

We work every day to find exceptional founders before consensus capital sees them, support them before the company is obvious, and be right often enough for the power law to do its work.

That is our job.

🎚️🎚️🎚️🎚️ Producer’s Note

Dvořák wrote it after arriving in America. A European composer, listening to a new world, without pretending he had stopped being European.

From the new world.

Best,

Fabian Hediger
Managing Partner
LinkedIn | Insta | X

Prof. Dr. Andy Ziltener
Managing Partner
LinkedIn | Google Scholar | Research Gate

StudioAlpha Capital is a Delaware-structured pre-seed venture fund backing AI-native B2B software startups at day zero. Legal counsel: Cooley LLP. Fund administration: AngelList.

We work with Cooley LLP on fund/legal structuring and AngelList on fund administration and investor onboarding.