Oren Zeev: 50% of Funds Will Go Out of Business

Oren Zeev manages over $2.7 billion across 11 funds as one of venture’s most prominent solo capitalists. His core thesis: radical alignment with LPs, zero personal income from management fees, and the conviction that being contrarian plus being right is the only recipe for great outcomes. In this conversation with Harry Stebbings, he systematically dismantles several popular venture narratives while offering a framework for navigating AI-era investing that prioritizes substance over appearance.

The Episode

This is a wide-ranging conversation between two investors with different vantage points. Harry Stebbings, running a multi-GP fund platform, pushes Zeev on the tensions he feels as an investor in the current market: inflated Series A pricing, the pressure to chase hyper-growth, the opportunity cost of not being in Cursor or Harvey. Zeev, operating as a solo GP with 30% carry and 13-14% personal LP commitment in every fund, pushes back on nearly every premise. The result is a surprisingly candid discussion about how incentive structures shape investment behavior, why most industry wisdom is wrong, and what the AI wave actually means for software incumbents.

Contrarian Investing in a Crowded Market

Zeev is firmly in the Peter Thiel camp: if everyone is doing something, that’s a reason not to do it. When Harry notes that every company now has 8-10 competitors at minimum, Zeev’s response is simply to avoid those situations entirely.

“If they look weird and they look wrong, then probably aren’t going to be 15 or 20 or 100 other startups doing it. So you’re probably going to have two or three years without real competition, and you have a chance of really building a real moat.”

His filter hasn’t fundamentally changed with AI, but he’s added one mandatory question: is this company a likely beneficiary of AI? If the answer is even neutral, it’s probably a no. Four years ago he wouldn’t have asked this question. Now it’s non-negotiable.

The Incumbents-Will-Die Narrative is Mostly Wrong

Zeev pushes hard against the popular “AI kills all incumbents” thesis, calling it a narrative promoted mainly by people whose motivation is to make provocative statements and get attention as thought leaders.

His framework for which software companies survive: the more operationally complex a business, the more it depends on distribution, regulatory licenses, ecosystem integrations, and proprietary data, the harder it is to disrupt. Technology is 5% of it. And in the age of AI, who has the most data? The incumbents.

He uses Navan (his largest concentrated position) as the case study. Simple software with limited operational complexity is vulnerable. But a business like Navan, which sits at the intersection of travel operations, airline relationships, enterprise integrations, and massive data sets, is not just defensible but a huge AI beneficiary. Gross margins have improved dramatically as AI handles support, and the customer experience improvements are even more exciting.

“If you have a piece of software that’s fairly simple, then yeah someone can write it quickly. But the more operationally complex a business is, the more it’s about distribution, integration, data… technology is 5% of it.”

The Dangerous Obsession with Hyper-Growth

When Harry raises the concern that 1-to-5 million revenue growth isn’t enough to attract top-tier follow-on investors, Zeev is blunt: “I don’t buy that either.”

His argument: AI doesn’t change mathematics. 2^5 is still 32. A company doubling annually with healthy economics for five years will be 32x its current size. He shares a live example of a portfolio company at $20M ARR growing to $40M with healthy unit economics, where an investor passed because “100% growth isn’t enough.” Zeev thinks that investor is dead wrong.

The real danger, he argues, is when the obsession with growth drives companies toward unsustainable behaviors, including circular deals where two companies buy each other’s products to inflate revenue. No value is created, but perceived value is.

“I think this notion that only growth matters is a very dangerous one, and I’ve seen this movie many many times.”

He acknowledges exceptions: in winner-take-all markets like Uber vs. Lyft, you don’t have the luxury of growing healthy. But for most businesses, optimizing purely for top-line growth while ignoring everything else is a “disaster waiting to happen.”

The Great VC Shakeout

Zeev’s prediction: at least 50% of venture funds currently in market either cannot raise, aren’t sure they can raise, or are stalling to avoid testing the market.

The bifurcation is clear: you’re either a platform (Andreessen, Sequoia, Lightspeed) that brings capabilities smaller VCs cannot match, or you’re differentiated in the opposite direction (solo GPs, domain specialists). The traditional five-to-six person partnership without anything uniquely special is caught in a painful middle.

“On one hand you’re not as agile, it’s not the personal connection. And on the other hand, you’re not Sequoia. So you’re not going to get the very best deals.”

For LPs, the picture is also grim. There’s been a drought of liquidity for four to five years. TVPI (Total Value to Paid-In) numbers are unreliable because GPs have enormous latitude in how they report valuations. Zeev’s rule of thumb: the more secure a GP is about raising their next fund, the less motivated they are to inflate numbers. Sequoia has zero incentive to inflate. A struggling mid-tier fund has every incentive.

He predicts a potential reshuffling in 2026-2027 with a “tsunami of liquidity” from unprecedented IPOs (SpaceX, Stripe, Databricks), which could change the dynamic significantly.

Radical Alignment: Zero Management Fee Income

Zeev’s fund structure is, by his own account, unique in venture:

• 13-14% personal LP commitment in every fund (the largest single LP in each)
• 30% carry (vs. the typical 20%)
• Zero personal income from management fees: he charges a low management fee but reinvests 100% back into the fund
• No payout before LP returns: he doesn’t see a dollar until LPs get 100% of their money back

This structure eliminates the core misalignment he sees in large funds: when a $10B fund charges 2%, the GP has already locked in $2B in management fees at close (20% over 10 years). The carry, even if the fund doubles, is another $2B but not for seven or eight years. After discounting for time value, the management fees are often worth more than the performance upside. This incentivizes GPs to optimize for raising the next fund rather than maximizing returns.

Within larger partnerships, the misalignment goes deeper: individual partners are first and foremost managing their careers. A partner who backed a struggling company has every incentive to convince their partners to put more money in (rolling the dice, buying personal time) rather than admitting failure, even when killing the investment would be better for the fund.

No Rules, No Strategy Disclosure

Zeev’s approach to LP relations is refreshingly blunt: “I tell LPs I only have one rule, and that rule is that I have no rules.”

He deliberately doesn’t tell LPs his investment strategy, minimum ownership thresholds, or deployment pace. His reasoning: if he commits to a specific approach, he feels obligated to follow it even when circumstances demand flexibility. He’d rather explain unusual decisions after the fact than be constrained in the moment.

He illustrates this with The Cart (an AI company): he deviated from his normal ownership targets, invested $1.5M for roughly 5% via a capped SAFE (he normally doesn’t do SAFEs at all), because the founders were exceptional. The company became profitable quickly and never needed more capital, so he never got to increase his position. But the investment has worked out well.

On deployment pace, LPs have complained that he invests too fast (sometimes deploying an entire fund in 12 months or less). His response: “I’m going to do my thing and if it works for them, fine. And if not, they can opt themselves out.” Some have opted out, and he’s fine with it.

Lessons from 2021 and the Price of Being Wrong

Zeev is candid about his 2021 mistakes: every deal he did was priced 3-4x what it should have been because “that was the market.” One of his funds (invested at peak) will be “okay but not great.” He also acknowledges having two “bubble-size” funds over $500M from 2021-2022, which he’s since corrected by cutting his latest fund to roughly $250M.

His broader philosophy on mistakes, heavily influenced by Annie Duke’s “Thinking in Bets”: don’t judge decisions by outcomes. In a probabilistic game like venture, you can make the right decision and still lose. The key insight:

“We humans are not truth seekers. We are self-validation machines.”

He applies this even to his own experience. An LP once pushed him to reduce fund size, delivered the feedback poorly, and Zeev refused to listen. In hindsight, the LP was right. Zeev’s takeaway isn’t about fund sizing but about how advice delivery matters as much as content, drawing a parallel to a parenting book about “how to talk to children so that they listen.”

Series A is Always Overpriced (and Always Has Been)

Zeev agrees that Series A is a difficult insertion point but insists this isn’t new. It’s always been the case that companies jump dramatically in perceived value between seed and Series A with minimal actual risk reduction. His advice: ignore the round label entirely and focus on whether there’s genuine signal of product-market fit, not just “now we have a product and a few logos.”

On preemptive rounds where a platform fund shoves $50M in shortly after a raise: his advice to founders is “take the money but continue to behave as if you didn’t.” Overfunding leads to loss of focus. If the founder is mature enough to park the money and spend based on market signals rather than boardroom pressure, they should take it.

A Few Observations

This conversation stands out for Zeev’s relentless focus on incentive structures as the explanatory variable for almost everything in venture: why GPs inflate valuations, why partners throw good money after bad, why LPs focus on DPI, why founders don’t listen to advice.

• The misalignment framework is the most valuable takeaway: every time you see a GP decision, ask what they’re optimizing for. The management fee math on large funds ($2B guaranteed vs. $2B in carry after 8 years) explains more industry behavior than any investment thesis.
• His “AI beneficiary” filter is deceptively simple but powerful. The nuance isn’t just “does AI help this company” but “is the competitive moat built on things AI can’t replicate” (data, operations, regulation, distribution). Pure software plays are vulnerable; operationally complex businesses are not.
• The 50% fund mortality prediction is probably conservative. If LPs continue to concentrate allocations in platforms while demanding DPI in a low-liquidity environment, the squeeze on mid-tier funds will only intensify.
• His philosophy of “no rules” and refusing to disclose strategy to LPs is an extreme version of what many solo GPs practice implicitly. It only works because of the radical alignment structure (largest LP, zero management fee income, no payout before LP returns). Without that structure, the lack of transparency would be a red flag.