Episode summary
Why AI Has Made Venture Capital Harder (Not Easier)
How I Invest With David Weisburd
· · 1 hr 15 min · Hosts: David Weisburd · Guests: Michael Gilroy
Summary
Michael Gilroy argues that AI has made venture capital harder because it has lowered the barriers to starting companies while flooding every trend with more founders, more capital, and more lookalike products. That creates more excitement at seed and pre-seed, but also makes diligence more demanding: investors can no longer rely on transcripts, polished demos, or revenue alone to tell them whether a company has real pull.
His framework starts with founders, but he says the real work is trend, TAM, and business model. He prefers to size markets from market cap down, then stress-test the underlying end market, customer behavior, and eventual earnings power. In AI, he thinks the biggest mistake is backing companies that are bought out of fear rather than need, or that are the number two, three, or four player in a crowded category.
Gilroy also argues that the best AI companies will be split between “capital incinerators” and “capital savers,” and that the winners will often be mature, disciplined founders with the poise to ignore tech Twitter and keep building. Marathon Management Partners is built around a small team, heavy human interaction, and concentrated ownership, with Gilroy emphasizing that the firm tries to win deals through long-term value creation rather than speed or brand halo.
Key takeaways
- Gilroy says AI has made venture harder because there are now 20 companies in a trend instead of two or three, and barriers to entry are “completely gone.”
- He underwrites deals around trend, TAM, and business model, and wants software that can trade at a premium to NASDAQ multiples at scale.
- He thinks revenue quality matters more in AI because buyers are often driven by fear and FOMO rather than durable need.
- Gilroy divides AI startups into “capital incinerators” and “capital savers,” and says mature, capital-efficient founders are the most convincing.
- Marathon’s edge, in his view, comes from concentrated focus, direct GP access, and doing a small number of deals with heavy diligence and customer calls.
Notable quotes
“AI has made it easier than ever to start a company, but according to today's guest, that might be making venture capital harder, not easier”
“We have instead of two to three companies in every trend, we have 20. And so this makes picking early extraordinarily difficult, but also a lot more exciting”
“Revenue quality is much much lower if somebody's buying it out of fear instead of out of need”
“We don't even talk about team because for us that's a given”
“The biggest mistake you can make in our business is backing the number two, three, or four business in any given market that ends up working”
How the episode opens
AI has made it easier than ever to start a company, but according to today's guest, that might be making venture capital harder, not easier. Michael Gilroy is the co-founder of Marathon Management Partners, which has $400 million across software and fintech. Michael, your view is that AI has changed venture capital fundamentally. What has AI changed exactly?
AI is very exciting. Everything we're looking at today is AI related or AI adjacent in some capacity. We like to take a macro lens on everything that we do.
A few things are happening on the company building side. One, we have more infrastructure than ever before. You kind of look at the last 10 years.
What did it take to start a company, right? And there's all this tooling now across general infrastructure. We go back to even the Twilio era of messaging and then today in AI, you can stand up a company really overnight.
The second component to starting a company are IPO uh potential style founders. So we have Google, Facebook, Amazon and then the new crop of large uh businesses, OpenAI, Anthropic, etc., etc. spitting off this talent at a record pace.
And so what does this mean for seed and preede investors? We have instead of two to three companies in every trend, we have 20. And so this makes picking early extraordinarily difficult, but also a lot more exciting.
Particularly when we're seeing companies go public in the north of a trillion dollar market cap range,
Barriers to entry are essentially gone. — Gone. Completely gone. You have an idea, you can spin up the product overnight. We have a joke internally, especially when we're talking to IPS and and we're talking about this uh exact phenomena. If even in this room, if the four of us in this room had an idea right now, we can be out selling it in probably a week from now.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- founder-first investing
- venture capital decision making
- ai changing vc landscape
- startup boards
- investor conviction vs consensus
Companies and tickers mentioned
- MSFT
- Coatue
- Microsoft
- M12
- Battery Ventures
- Insight Partners
- Marathon Management Partners
Watch the episode
Watch “Why AI Has Made Venture Capital Harder (Not Easier)” on YouTube
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from How I Invest With David Weisburd
-
Hunter Walk: Why Venture is Getting Harder, Picking Great Founders, and Investing Against Consensus -
True Anomaly’s Even Rogers on Space Warfare, China’s Military Advantage, and the Defense Tech Boom -
Finding the Next Anthropic, Quantifying Venture Capital, and the Science of Picking Winners -
Why 95% of AI Projects Fail, Anthropic still at $2T?, and Why AI Is Still in the First Inning -
Peter Thiel's 27-Year Partner on the AI Bubble, the Venture Capital Crisis, and the PayPal Mafia -
Why Venture Is Eating Public Markets, Picking Great VCs, and the AI Boom w/Brijesh Jeevarathnam
How I Invest With David Weisburd, this episode and its artwork belong to David Weisburd. Octagon is not affiliated with the show; this page is an independent summary.
