Why It's Hard for Small Venture Funds to "Play the Game on the Field" in AI Investing

I'm not sure the analogy of playing the game on the field works for small funds who are thinking about how to navigate the world of AI.

The venture capital industry tends to move in cycles, and sometimes, it’s useful to go back and read perspectives written in previous cycles. One of the posts I often reference is this post about bubbles by Bill Gurley from Benchmark. There is also a great podcast interview he did with Kara Swisher that expands on some of these points. The TL;DR is that all companies exist in an environment where their behavior and the behavior of other related parties (investors, employees, competitors) are influenced by all parties’ collective belief as to how the world works at that moment in time in terms of the availability of capital, the balance between profitability and growth and other market forces. I am not suggesting that AI is a bubble; I am pretty convinced that AI will be a transformative technology, but I have many open questions about how the gains from AI advancement will be split between the producers and consumers of AI technology (your classic econ analysis about who captures the surplus) and whether startups or incumbents will capture more of the value on an industry-by-industry basis. Nonetheless, when I re-read those posts and talk to my peers today, I think the “game on the field” analogy is very important in understanding how the venture business works today.

Most small, early-stage venture capital firms whose funds are less than $250 million in size are not set up to fully finance their pre-seed and seed-stage companies from the first check to IPO. The venture capital industry has historically worked as a relay race where investors at one stage bring in the next stage of investors to fund and support companies as they scale. Right now, it feels like AI investing breaks this model in a few important ways:

• The largest venture capital firms are laser-focused on opportunities that can produce $5-10B minimum outcomes. If you manage a multi-billion dollar venture firm, you need exits of that scale to make your fund math work. Right now, the consensus belief is that enterprise and B2B-focused AI-powered companies are most likely to produce outcomes of that scale. Some notable exceptions exist for firms focused on deep tech, hard tech, defense tech, or other themes and are not almost exclusively focused on AI. I think those firms are notable because they are exceptions to the rule.
  

• Many founders and founding teams with experience building AI-powered products at scaled companies can raise money from multi-stage venture capital funds with round sizes and valuations that are outside the remit of what most smaller venture funds can afford to pay given their portfolio constructions, fund sizes, and strategies. In some cases, seed funds won’t see these teams at all as they will go directly to multi-stage firms for their first rounds of capital. In other cases, seed firms will see these rounds but get priced out as the multi-stage firms make offers those seed firms cannot and probably should not match.

• Companies with strong early-stage traction in AI can get follow-on financing quickly because so much of the energy in venture right now is looking for those companies. If you are fortunate or skilled enough to be an investor in an early breakout AI company, you can get follow-on financing that will mark up your position quickly and aggressively.

If you are a small venture fund, this creates quite a conundrum:

• If you don’t invest in AI and continue to invest in other categories, you might end up with companies you believe in but have a very narrow or nonexistent market for follow-on financing. Right now, most of the largest funds are focused on AI. Getting anyone to pay attention and care about that company might be hard if you have companies performing well but are not leveraged to an AI thesis. That creates follow-on fundraising risk, and no seed investor or early-stage founder can control the preferences or interests of the next investor. We live in a world where most smaller funds depend on larger, multi-stage funds to invest in the next round to keep their companies going. An added wrinkle is that an investment made in seed today has to be interesting to the venture market that will exist 18-24 months from now; nobody knows what that world will look like, and there is a strong temptation to extrapolate today’s market into the future.

• Many founders with domain expertise are not raising small pre-seed and seed rounds; they are going directly to multi-stage funds and raising larger rounds of capital at higher valuations for their initial rounds. If you see these rounds and have the opportunity to invest, the prices and associated ownership will likely break your portfolio construction, and these companies will look like exceptions to the model. It’s hard to build a portfolio of companies that are all exceptions to the model you pitched your limited partners when you raised your fund.

• We are still a ways away from figuring out where the long-term true value creation and value capture will happen when it comes to AI investing. That being said, the enthusiasm for AI could persist for some time, and there are seed fund managers who might need to raise new funds before the market answers some of these key questions about value capture and value creation. Having a portfolio on the wrong side of that theme (either being long on AI when the market has decided it’s less interesting or being light on AI when it’s all the rage) has real implications for the ability to raise the next fund. So there’s that added wrinkle, too.

I write this post with full knowledge that I don’t have the answers and I am trying to figure this all out as we go. I don’t remember the time I saw a technology shift that was this impactful and unpredictable simultaneously. The venture industry is in for quite the ride with AI.