Our 2Q letter to LPs | 99d Summary

1. Key Themes

Theme 1: We Are in the "Frenzy" Phase of the AI Technology Cycle — and That Changes What's Investable

Rechtman maps current AI investment onto a well-worn historical pattern of technology bubbles and crashes, warning that today's euphoria is legible, not exceptional.

"Every general-purpose/horizontal technology charts a familiar course between early excitement and mature, scaled deployment... Eventually every bubble burst (railways in 1847 and 1873, utilities in 1929, fiber in 2001, crypto in 2018) leaving behind critical infrastructure for a next generation of companies and investors to build newly possible companies. Clearly we are in 'The Frenzy' today."

Theme 2: The Supply-Side Argument for AI Is Not Sufficient for Long-Term Value Capture

The "it's different this time" claims about AI — speed, scale, impact on knowledge work — are supply-side claims and don't determine who captures lasting value.

"Value capture is not principally gated by the supply side (the vendors). It's gated by the customer prerogative: how long it takes a hospital, a bank, a law firm, etc. to refactor around the new tool. A better model doesn't buy you a faster customer because the binding constraint doesn't move."

Theme 3: Consumers of Technology, Not Developers, Historically Capture the Most Value

The long-term thesis is rooted in a consistent historical pattern: the biggest winners are downstream users of transformative infrastructure, not the builders of it.

"Historically, the biggest long term winners have always been the consumers and users of a technology, not its developers: Steel producers and oil majors vs cars / Internet service providers vs internet platforms / Electric utilities vs electrified industry / Datacenters vs AI, ironically."

"The market cap of steel today is a fraction of the market cap of planes and cars which is a fraction of the market cap of all the businesses made possible by planes and cars."

Theme 4: The Investment Target Is "Newly Possible" Businesses That Sell Outcomes, Not Workflow

The fund's specific focus is not AI infrastructure but the businesses AI makes newly viable — particularly those that can shift from selling software processes to selling results.

"We want to back things that are newly possible using AI infrastructure, talent, and technology which also conform to our ideas about what makes a good business (terminal value)."

"Increasingly, we think/see/can prove that AI brings businesses closer to selling outcomes not workflow. The best businesses have always sold outcomes and now there's more surface area to do that in big markets."

2. Contrarian Perspectives

Contrarian 1: "AI Will Change Everything" Is Not, By Itself, an Investment Thesis

The consensus treats AI's economic disruption as a sufficient reason to invest broadly. Rechtman explicitly rejects this.

"We are long-term bulls. AI is going to break (is breaking) the economics of industries everyone assumed were untouchable... Much of the market treats that as a sufficiently investable idea unto itself. It is not, at least not for us."

"Accepting that the economics break doesn't tell us enough about where to put the next dollar. The real questions come after: Who is organized to rebuild the broken economics and win? Where does the value sit long term? Who can defend it once they've got it?"

Contrarian 2: AI Infrastructure Builders (Including Data Centers) Are the Losers, Not Winners, of the AI Cycle

Against the prevailing excitement around picks-and-shovels AI plays (GPU makers, cloud providers, model companies), Rechtman puts them on the losing side of the historical analogy.

"Datacenters vs AI, ironically" — listed alongside ISPs vs platforms and utilities vs electrified industry as examples of infrastructure providers who lost to downstream users.

This implies that the market's current emphasis on Nvidia, hyperscalers, and frontier model companies may be misallocating long-term capital, consistent with historical patterns where infrastructure mania preceded the actual value-creation era.

Contrarian 3: Adoption Speed Is Constrained by Customers, Not Technology — Making "Fast Diffusion" Arguments Irrelevant

The common bull case for AI is that it moves faster than prior tech transitions. Rechtman argues this misses the binding constraint entirely.

"The cycle of diffusion is slow, even if its onset is sudden... What's different specifically — the scale, the speed, the sophistication, that it's finally coming for knowledge work? All in some sense true but orthogonal to the point of how businesses work and who 'decides.'"

The facts: hospitals, banks, and law firms operate on institutional timelines that no model improvement can accelerate — making technology velocity a weak predictor of enterprise revenue ramp.

3. Companies Identified

Company	Description	Why Mentioned	Quote
No specific companies named	—	The letter is thematic/framework-oriented, not company-specific	—

Note: The article references Carlota Perez's framework and historical industries (steel, oil, railroads, fiber) as illustrative analogies, but names no current investable companies.

4. People Identified

Person	Description	Why Mentioned	Quote
Yoni Rechtman	Partner at Slow Ventures, leads pre/seed rounds from ~$325M fund	Author; articulates the fund's AI investment framework	"I'm a generalist investor looking for weird takes on important stories: N-of-1 companies taking non-obvious approaches to markets that matter."
Carlota Perez	Economic historian, author of Technological Revolutions and Financial Capital	Her framework for technology frenzy/crash cycles underpins the entire investment thesis	Referenced via chart: "Carlota Perez: Technological Revolutions and Financial Capital"
Tomasz Tunguz	Venture capitalist, writer	Linked for further reading on "Valuing AI Harnesses"	Listed in "Elsewhere" reading section
Phin Barnes	Investor/writer	Linked for further reading on "It's time to build software"	Listed in "Elsewhere" reading section
Nathan Lee	Investor/writer	Making a case for consumer pre-seed investing that Rechtman is actively debating	"Nathan and I have been debating this for the last couple weeks. I'm not completely sold yet but he thinks he's onto something."
Vincent Bevins	Journalist/author	Linked for cultural/macro piece on the "vibe shift" as leveraged buyout	Listed in "Elsewhere" reading section

5. Operating Insights

Insight 1: Build Businesses That Sell Outcomes, Not Workflow — AI Makes This More Achievable Than Ever

For founders, the strategic implication is clear: structure your value proposition around the end result the customer gets, not the process or software tool you provide. AI expands the surface area where this is viable.

"Increasingly, we think/see/can prove that AI brings businesses closer to selling outcomes not workflow. The best businesses have always sold outcomes and now there's more surface area to do that in big markets."

Insight 2: Time to First Revenue Is Not a Reliable Signal of Future Growth — So Don't Optimize for It

Founders and early-stage operators often treat early revenue milestones as validation of trajectory. Rechtman's data suggests this is a weak predictor.

"Consistent feature, kind of surprising, worth remembering: time to first revenue is rarely a meaningful predictor of future growth rates. But once you get going you gotta really go!"

The implication: prioritize getting the model right over getting the first dollar fast — but once revenue begins, growth velocity becomes the critical variable.

6. Overlooked Insights

Overlooked Insight 1: The Three Questions That Should Gate Every AI Investment

Buried in the broader framework argument is a crisp investment checklist that applies equally to founders stress-testing their own positioning:

"Who is organized to rebuild the broken economics and win? Where does the value sit long term? Who can defend it once they've got it and what does it cost them to hold? Everything short of answering those is a trade."

This is a durable filter for separating durable businesses from momentum plays — and most AI companies today likely cannot fully answer all three.

Overlooked Insight 2: The Fund Is Now Accessible via MCP Server — A Novel LP/Founder Interface

Barely mentioned and easy to miss: Rechtman notes that "you (or your agents) can read/chat with 99D via my MCP server." This is a quietly significant experiment — a VC making their thesis and content programmatically queryable by AI agents, potentially signaling a broader shift in how investors will interface with founders and deal flow in an agentic world.