Key Takeaways

  • Jack Altman argues the traditional $3M to $6M seed round buying 8% to 15% ownership is evaporating as top founders leap straight into massive AI rounds.
  • Jason Lemkin points out that $2M to $3M remains the atomic unit of capital required to build software, yet building slowly no longer protects you from platform risk.
  • Rory O'Driscoll explains that the classic bargain of slow compounding (lower growth in exchange for lower risk) is broken because AI creates high obsolescence risk across slow-moving software.
  • O'Driscoll estimates that the venture market is now divided into an 80% fast-action arena dominated by rapid AI capital and a 20% slow-action bracket for traditional software.

The Disagreement

For a decade, early-stage venture operated on a predictable rhythm: raise $3M to $6M, prove product-market fit, and compound at 80% year-over-year. Altman argues that this middle ground has collapsed. In high-demand lanes, founders skip standard seed mechanics entirely. As Altman put it: “In many lanes, I think that round has kind of evaporated. And so I think there is a cohort of the market where traditional seed investing where you're going to write 3 to $6 million checks by you know 8 to 15% where I just think that is fully broken slash just isn't there anymore.”

Lemkin defends the capital math while acknowledging the existential risk. He maintains that modern cloud infrastructure and developer tools mean a lean team can still launch a real business on modest funding: “The truth is you can do as much I think for 2 to three million bucks as you could 10 years ago. And if you don't have folks dying to give you capital outside of demo day that that may still be the natural atomic amount of capital.” But Lemkin admits the outcome profile has soured: “The problem with the quiet compounder, to Jack's point, in my view, as someone who's pitched quiet compounding since 2012, is they're just not with exceptions. They're just not stable.”

O'Driscoll targets the underlying risk-return ratio. Slower vertical software businesses used to offer safety in exchange for modest growth rates. Now, rapid model releases threaten to make slow-baking workflows obsolete before they reach scale. As O'Driscoll stated: “If you're making a compounding play, the quid proquo should be low risk. And if the world is such that the tech environment is changing so much that you get the compounding, not the hyperrowth, but you get the same level of risk, that by definition is a sub-optimal game.”

Who's Right (and When They're Wrong)

O'Driscoll and Altman are right about venture math, while Lemkin is right about operating reality.

If you are pitching institutional venture funds, the quiet compounder is an unattractive bet. O'Driscoll pointed out that “the table at the moment is 80% the fast action table and 20% the slow action which makes sense because in 2022 there was a discontinuity and everything before that became obsolete.” When platform risk is high, taking five years to hit $5M in annual recurring revenue means your feature set might be absorbed by an upstream model before your Series B.

However, Altman's view fails for founders outside the tier-one AI spotlight. If you cannot command a $30M pre-seed valuation from day one, attempting to play the hyper-speed game will leave you stranded when momentum stalls. Lemkin's atomic unit of $2M to $3M remains the only viable path if you own deep domain workflows that foundation models cannot easily ingest.

What to Do With This

Audit your product roadmap this week against foundation model releases. If your three-year growth plan relies on slow enterprise compounding, identify the one proprietary dataset or workflow integration that an upstream model cannot replicate next quarter. If you cannot name it, compress your development milestones from twelve months to ninety days to test whether your product survives current platform shifts.