There has rarely been a more confusing time to be a venture capitalist.
On one hand, software has never felt more magical. A solo founder, and increasingly even a non-technical one, can use AI tools to prototype, design, code, test, deploy and iterate an entire product in days. GitHub’s research on Copilot showed developers completing a coding task 55% faster with AI assistance, and Stack Overflow’s 2025 Developer Survey found that 84% of respondents are already using or planning to use AI tools in their development process.
On the other hand, the bar for building a venture-scale company has rarely been higher.
It is easier than ever to create an app. It is harder than ever to make anyone care. It is easier than ever to write code. It is harder than ever to build something defensible. It is easier than ever to launch. It is harder than ever to stand out when thousands of other founders can launch something similar at roughly the same time.
This is the paradox VCs are wrestling with in 2026. The tools are better, talent is more leveraged, the market is bigger, and yet the signal is noisier, capital is more concentrated, and the definition of “exceptional” keeps moving.
AI made software easier. It did not make company-building easier.
AI coding tools are real. They are not just a productivity narrative. Every founder now has access to what feels like a small engineering team in a browser tab. The distance between idea and demo has collapsed.
But that superpower is available to everyone.
That changes the venture equation. If everyone can build faster, then speed of building alone stops being a moat. The question shifts from “can this team build?” to “can this team build something that compounds?”
That means investors are spending more time asking harder questions. Is there proprietary data? Is there deep workflow insight? Is the founder solving a painful enough problem? Is the product embedded in a customer’s day-to-day behavior? Can the company acquire users efficiently? Can it retain them? Does the AI improve with usage? Is there technical IP or domain expertise that cannot be replicated by a prompt and a weekend?
The more powerful the tools become, the more important taste, insight and distribution become.
When everyone sees the same opportunity
Another challenge for VCs is that AI makes certain problems visible to everyone at the same time.
Take agent reliability, monitoring of agent outputs, AI security, enterprise copilots, code review, AI customer support, synthetic data, vertical AI assistants or workflow automation. Once a new platform shift becomes obvious, multiple talented teams often start companies around the same problem within months of each other.
Sometimes that is a sign of a real market. If several smart founders are independently attacking the same pain point, there is probably something there. But it also creates a difficult question for investors: how many of these companies can actually become large outcomes?
In some categories, the market may support one or two major winners and a long tail of smaller companies. In others, the feature may get absorbed by incumbents, cloud providers, model companies or larger platforms. And when the underlying technology is not proprietary, the investor has to ask whether the company is building a product, a feature, or simply a temporary interface on top of someone else’s model.
This is where execution and go-to-market become decisive.
If there is no obvious technical moat, the company that raises the most capital, hires the strongest team, signs the best early design partners and gets in front of customers fastest can gain an advantage. Venture capital can become a form of kingmaking. The best-funded company does not always win, but funding can help create the conditions for winning: better talent, more enterprise credibility, faster sales coverage, stronger partnerships and more time to iterate.
That creates another uncomfortable dynamic for VCs. The decision is not only “is this a good company?” It is also “is this the team that can win this market?” and, sometimes, “will our capital help make them the winner?”
This is especially hard at seed, when evidence is incomplete. Several companies can have similar decks, similar demos, similar customer conversations and similar visions. The real differentiation may come down to founder quality, speed of learning, depth of customer insight, credibility with buyers, and the ability to recruit exceptional people before the market becomes crowded.
In a world where many products can be copied, the company itself becomes the moat. The team, the speed, the customer intimacy, the distribution, the brand and the ability to keep compounding are what separate the eventual winner from the many credible attempts.
The SaaS playbook is being rewritten in real time
For the past 15 years, venture investors had a fairly well-understood SaaS playbook. Find a painful workflow. Build a cloud product. Sell to an ICP. Track ARR, retention, CAC payback, gross margin and expansion. The best companies could become category leaders with strong net revenue retention and high gross margins.
That playbook is not dead, but it is under pressure.
Carta’s 2025 data shows the contradiction clearly. Series A deal count on Carta was down 18% year-over-year in Q2 2025 and cash raised fell 23%, even as median Series A valuations hit a new high of $47.9 million. In other words, the market is not closed. It is selective. [2]
The result is strange: you can meet a very good SaaS company today, with revenue, customers and a competent team, and still worry that it may not be “venture obvious” enough.
That phrase, “venture obvious”, is part of the problem. In a capital-rich consensus market, companies often need to look inevitable before they are fundable. Growth needs to be faster. The market needs to feel larger. The wedge needs to be sharper. The AI story needs to be credible. The team needs to have a right to win.
A few years ago, a horizontal SaaS company growing steadily could attract strong investor interest. Today, unless it is growing very fast, has a clear AI-native advantage, or owns a valuable niche with real depth, it risks being seen as “good, but not enough”.
That is brutal for founders. It is also challenging for investors. The talk about SaaSpocalypse isn’t helping investors get more confident, either.
Capital is concentrating at every level
The venture market is not behaving like one market. It is behaving like several markets at once. Just in Q1 2026, over $300 billion VC dollars were deployed into startups globally. The graph below shows what this means in context. But while the total amount deployed has catapulted, the number of deals is actually declining. So less companies are attracting more capital.
At the fund level, capital is flowing to a small number of established platforms also. PitchBook-NVCA reported that in Q1 2026, 73.1% of capital committed went to just five VC firms. In the same quarter, experienced managers captured more than 90% of capital raised. This had a pretty massive impact on the number of venture capital funds. It dropped from 1,609 in 2022 to 537 in 2026.
At the company level, capital is also concentrating in fewer companies and sectors. In Q1 2026, PitchBook-NVCA reported that $195.6 billion was invested in just five companies. The same report said 88.8% of Q1 deal value went to AI and machine learning companies, while 42.5% of deal count involved an AI startup.
Globally, the OECD found that AI companies attracted 61% of all venture investment in 2025, or $258.7 billion out of $427.1 billion. The United States alone accounted for about 75% of global AI VC deal value.
This is why the market feels so strange. The headline numbers can look euphoric while the lived experience for many founders and funds feels constrained. Mega-rounds make the market look liquid. Emerging managers struggle to raise. AI labs and infrastructure companies attract enormous capital. Good non-AI companies can find themselves fighting for attention. Silicon Valley absorbs more of the narrative and the dollars, while founders outside the main hubs have to work harder to be seen.
The average does not describe the market anymore. The median probably does not either. Venture is increasingly defined by the outliers.
The allocation problem for VCs
For VCs, this creates a real allocation dilemma.
If you ignore AI, you risk missing the most important platform shift in technology. If you chase every AI deal, you risk paying inflated prices for companies with shallow moats. If you only back companies with revenue, you may miss the deeply technical teams building breakthrough infrastructure. If you only back deep tech, you may miss the consumer and application-layer companies that turn new capabilities into daily behavior.
The classic venture tension has become more acute: should you back what is obviously working, or what is still non-consensus but could matter more?
There is also a portfolio construction question. In a market where capital concentrates around perceived winners, the temptation is to pile into consensus deals. But consensus often comes with high prices, compressed ownership and crowded cap tables. The alternative is to invest earlier, with more uncertainty, before the market has decided. That is emotionally harder, but potentially more rewarding.
The challenge is that “early” has changed too. A pre-seed company today can have a polished product, paying customers, AI-generated demos, a global waitlist and a team of two. At the same time, some of those signals are easier to manufacture than ever. The demo can be misleading. The revenue can be pilot-heavy. The product can be a thin layer on top of someone else’s model. The growth can come from novelty rather than retention.
For investors, diligence has to go deeper. The question is not just “does it work?” It is “will this still matter in 18 months?”
Growth or depth
In this market, I increasingly look for one of two things: unusual growth or unusual depth.
The first is speed. Some companies grow so quickly that they become impossible to ignore. They find a wedge, capture demand, and show week-over-week or month-over-month momentum that suggests something unusual is happening. In consumer, that might mean retention and organic sharing. In B2B, it might mean fast revenue, expansion and clear ROI. In AI, it might mean usage that compounds because the product becomes more useful with more data or more workflow integration.
The second is depth. Some companies have a technical insight, domain expertise, proprietary data, regulatory advantage or workflow intimacy that is genuinely hard to copy. They may not look explosive on day one, but they are building from a foundation that can become defensible over time.
The hardest companies to fund today are the ones in the middle: not growing fast enough to stand out, not deep enough to be defensible, not cheap enough to be obvious, and not differentiated enough to overcome investor hesitation.
That does not mean they are bad companies. Many may become profitable, durable, valuable businesses. But venture capital is not designed to fund every good business. It is designed to find the few companies that can return the fund.
That distinction matters more now.
The opportunity is still enormous
Despite all of this, I remain optimistic.
In fact, the confusion itself may be the opportunity. When markets become noisy, conviction matters more. When capital crowds into obvious categories, overlooked areas can become attractive. When tools commoditize execution, founders with deep taste, insight and lived experience become more valuable. When everyone can build, knowing what to build matters more.
At Remagine Ventures, we continue to get excited by founding teams that combine ambition with depth. We are especially interested in Israeli founders globally using AI to reimagine how people spend their time and money across entertainment, gaming, media, commerce and consumer behavior.
We like founders who understand a market from the inside. Founders with a non-obvious wedge. Founders who can move fast, but are not just chasing the trend of the week. Founders who have a reason to win beyond “we can build this now with AI”.
This market is confusing because two things are true at once. It has never been easier to start. It has rarely been harder to break through.
For founders, that means the bar is higher. For investors, it means the job is harder. But for the very best teams, the opportunity remains as exciting as ever.
The next great companies are still being formed. They may just look different from the last generation. They may be smaller, faster, more technical, more creative, more global and more AI-native from day one.
And that is exactly what makes this moment worth paying attention to.
He is a former General Partner at Google Ventures (GV) in Europe, former head of Google for Entrepreneurs in Europe, and founding head of Campus London, Google's first startup hub. Eze writes on Israeli tech, venture capital, artificial intelligence, and founder strategy.
He is also the founder of Techbikers, a nonprofit that brings together the startup ecosystem on cycling challenges in support of Room to Read.
- The Venture Capital Paradox: Why This Market Is So Hard to Read - June 24, 2026
- Geopolitics Fuels Israel’s $12bn Defense Tech Surge - June 22, 2026
- Weekly Firgun Newsletter – June 19 2026 - June 19, 2026
