This is the forward-looking module, and it comes with a caveat: predicting venture's future is a good way to look foolish later. So rather than forecasts, this module maps the forces visibly reshaping the industry — the AI wave, the liquidity revolution, the flood of capital, the shifting geography, and the new fund structures — and frames the genuinely uncertain parts as open questions rather than predictions. The goal is to equip you to reason about where venture is heading, not to hand you answers that will date badly.
Before mapping the forces reshaping venture, a word of intellectual honesty. The history of confident predictions about venture capital is mostly a history of being wrong. People who declared venture dead after the 2000 dot-com crash missed the social and mobile booms that followed. People who declared in 2021 that valuations would only go up were brutally corrected in 2022. The industry is shaped by technological waves that are inherently hard to predict, and by macro conditions (Module 12) that move on their own schedule.
So this module is built differently from the others. Instead of teaching settled material, it identifies the forces that are clearly at work and reasons about their implications — while flagging, honestly, where the outcome is genuinely uncertain. The valuable skill is not memorizing a forecast; it's understanding the forces well enough to update your own view as events unfold.
Each section identifies a force that is observably reshaping venture — these are not speculative; the trends are real and documented through the mid-2020s. What's uncertain is where each leads, and those uncertainties are flagged in open-question boxes throughout. Treat the forces as facts to understand and the questions as genuinely open — places where informed people disagree and where the next few years will provide answers. A student who finishes this module should be able to follow venture news and slot new developments into this framework, rather than being surprised by them.
One more framing note: many of these forces interact. The AI wave affects capital flows; capital superabundance affects fund structures; the liquidity revolution affects the whole exit model. The sections below treat them somewhat separately for clarity, but the real picture is a system of interacting pressures, and the most important developments often happen at the intersections.
The most obvious force reshaping venture in the mid-2020s is artificial intelligence — and it operates on venture at three distinct levels, which are worth separating.
A large and growing share of venture capital has flowed toward AI companies — foundation-model labs, AI infrastructure, and applications built on top of AI models. This concentration is the most visible shift in what venture funds. It echoes prior platform shifts (the internet in the late 1990s, mobile around 2008-2010, cloud through the 2010s), where a new technology wave reorganized where capital flowed.
Foundation-model companies require enormous capital — for compute (GPU clusters), data, and talent — at a scale that strains the classic venture model. A traditional venture fund's check is small relative to the billions a frontier AI lab needs. This has pulled in non-traditional capital (sovereign wealth, corporate strategic investment from cloud providers, specialized infrastructure debt from Module 08) and raised the question of whether frontier AI is even a "venture" business in the classic sense, or something closer to capital-intensive infrastructure.
Venture firms are using AI in their own workflows — sourcing deals through AI-driven scanning of company data, automating parts of diligence, and analyzing portfolios. This is a quieter shift than the investment-theme change, but it may matter for the structure of firms: if AI can do much of the analytical work that junior staff once did, the shape of the venture firm (Module 09's career ladder) could change.
Does frontier AI break the venture model, or extend it? The capital intensity of foundation models is unlike anything venture has funded before — closer to building a semiconductor fab or a telecom network than a software startup. If the biggest AI companies need tens of billions in capital and behave more like capital-intensive infrastructure, the classic venture model (small early checks, power-law returns, equity for high-uncertainty work) may not fit them well. Alternatively, the application layer built on top of AI models may look like classic venture, with the infrastructure layer funded differently. Whether AI is a venture wave like prior ones, or a structurally different kind of business that venture only partly funds, is genuinely unresolved.
It's worth holding some humility here. Every major technology wave has prompted claims that "this time is different" and that venture must transform. Sometimes that's been true (the capital needs of AI infrastructure really are different); often the underlying model proved more durable than the wave-specific predictions. The honest position is that AI is clearly reshaping what venture funds and may be reshaping how, but whether it changes the fundamental model remains to be seen.
Module 11 introduced the growing separation between "exit" and "liquidity." This separation has become one of the defining structural shifts in venture, and its consequences are still unfolding.
The core dynamic: companies stay private far longer than they once did — fifteen or twenty years rather than five to seven — which created a liquidity problem for everyone holding their shares. The industry's response has been to build mechanisms that provide liquidity without a traditional exit:
When a venture fund reaches the end of its 10-year life holding a still-private winner, the continuation fund lets the GP move that position into a new vehicle, returning capital to the old fund's LPs (who can cash out) while letting the GP keep managing the position for new LPs who want continued exposure. This directly addresses the tension between the fund clock and the company's timeline that Module 09 described. Continuation funds raise their own questions — conflicts of interest (the GP is on both sides of the sale), valuation fairness — but they've become a significant part of the venture liquidity landscape.
If private companies can offer near-public liquidity, does the IPO still matter? The traditional reason to go public was partly to give shareholders liquidity. As secondaries, tender offers, and continuation funds increasingly provide that liquidity privately, the liquidity-driven motivation to IPO weakens — companies can stay private indefinitely while still letting stakeholders cash out. This could mean a permanently thinner IPO market and a much larger, more liquid private market. Or public markets could reassert their advantages (deeper liquidity, lower cost of capital, public-currency for acquisitions). Whether the future is "private companies forever, with private liquidity" or a return to the IPO as the natural endpoint is an open and consequential question.
Over the 2010s and into the 2020s, the amount of capital flowing into venture grew enormously — from the asset-class maturation (Module 09's endowment model spreading), from sovereign wealth (Module 04), from crossover funds, and from the general search for returns in a low-interest-rate world. This abundance of capital has structural consequences that may be the most important long-run force in the industry.
When more capital chases the same number of genuinely venture-scale opportunities, two things happen: valuations rise (more demand for deals) and average returns fall (the same winners spread across more invested capital). This is basic supply and demand applied to the asset class. The concern is that venture has attracted more capital than there are venture-scale opportunities to absorb it productively, which would mean lower returns for the asset class as a whole even as individual top funds continue to do well.
The dynamic interacts with the power law in an uncomfortable way. Venture returns depend on capturing a small number of enormous winners (Modules 04, 10). The supply of genuinely fund-returning companies in any given year is limited — it's a function of how many transformative companies the economy actually produces, which doesn't scale just because more capital is available. If capital doubles but the number of fund-returners doesn't, the returns per dollar invested must fall. This is the superabundance problem in its starkest form.
Has venture become too big to deliver venture returns? If the asset class has attracted more capital than there are venture-scale opportunities, then aggregate venture returns should decline toward more ordinary levels, even as the best funds continue to outperform. This would represent a maturation of venture from a high-return niche into a larger, lower-return asset class — more like private equity became. Alternatively, technological waves like AI might be expanding the universe of venture-scale opportunities fast enough to absorb the capital. Whether the future is "venture returns regress to the mean as the asset class matures" or "new technology waves keep expanding the opportunity set" is unresolved and matters enormously for everyone in the industry.
Module 13 mapped the world's ecosystems. The forward-looking question is how the geographic balance shifts from here. Several forces are pulling in different directions.
The China decoupling (Modules 04, 06, 11, 13) casts a long shadow over the geographic picture. The partial separation of the U.S. and Chinese venture ecosystems — once deeply interconnected through dollar funds and U.S. listings — represents a structural fragmentation of what had been a globalizing industry. The long-run consequence may be two somewhat separate venture worlds (a U.S.-aligned one and a China-aligned one) rather than the single global market that seemed to be emerging in the 2010s.
Does venture globalize further, or fragment into blocs? Two opposing trends are visible. On one hand, talent is more global, emerging ecosystems are maturing, and capital flows across borders more than ever. On the other, the U.S.-China decoupling has fragmented the largest cross-border relationship, AI concentration is reinforcing U.S. dominance, and geopolitical tension is making cross-border venture investment more fraught (more scrutiny of who invests in what, on national-security grounds). Whether the future is a more dispersed global venture market or a fragmentation into geopolitical blocs — with capital, talent, and companies flowing freely within blocs but not across them — is one of the most consequential open questions, and it depends heavily on geopolitics outside the industry's control.
The traditional venture firm structure (Module 09) — a partnership raising a series of 10-year funds from institutional LPs — is no longer the only model. A range of new structures has emerged, unbundling and reimagining how venture capital is organized.
"Solo capitalists" — individuals raising and deploying significant funds under their own name, without a traditional partnership — have become a real category. Some manage hundreds of millions of dollars alone. "Rolling funds" (a structure popularized in the early 2020s) let investors raise capital on a continuous quarterly subscription basis rather than in discrete 10-year vintages, lowering the barrier to starting a fund. These structures unbundle the traditional firm, betting that an individual with strong judgment and network can outperform a partnership without the overhead.
Sequoia's 2021 move to a long-duration evergreen structure (Module 09) was an early signal of a broader interest in escaping the fixed 10-year fund life. Evergreen and permanent-capital vehicles can hold positions indefinitely, recycle capital, and avoid the forced-exit pressure of the traditional clock — directly addressing the tension between fund timelines and companies staying private longer. More firms are experimenting with these structures.
At the opposite end from solo capitalists, the largest firms (a16z is the most-cited example) have evolved into something more like diversified financial-services platforms — multiple funds across stages and asset types, large staff, extensive "platform" services (Module 04). This represents a different bet: that scale, brand, and a full-service offering win, rather than the lean high-judgment model. The industry is bifurcating between these extremes — tiny solo funds and giant platforms — with the traditional mid-size partnership squeezed in between.
Does the traditional mid-size venture partnership survive the barbell? The industry appears to be bifurcating into a "barbell": lean solo capitalists and rolling funds at one end (low overhead, high judgment, nimble) and giant platform firms at the other (scale, brand, full-service). The classic mid-size partnership — a handful of partners raising a few hundred million per fund — is squeezed between them. Whether the mid-size partnership remains the industry's backbone, or gets hollowed out as capital concentrates at the extremes, is an open structural question. The answer matters because the mid-size partnership has historically been where much of venture's best returns came from (Module 10's Benchmark model).
A forward-looking module owes the reader an honest treatment of the serious critiques of venture capital — not because they're necessarily right, but because reasoning about venture's future requires engaging with the strongest arguments that something is wrong.
The superabundance argument (Section 04), taken to its conclusion: the asset class has attracted far more capital than the real economy can productively absorb in venture-scale opportunities, which means much of that capital is being misallocated — funding companies that shouldn't be venture-funded, inflating valuations, and depressing returns. On this view, venture is due for a significant contraction, and the 2022-2023 correction was a foretaste rather than a one-off.
A different critique: venture's power-law incentive structure (it needs billion-dollar outcomes) systematically biases it toward certain kinds of companies — those that can scale enormously and exit richly — and away from others that might be more socially valuable but can't produce venture-scale returns. Critics argue this distorts what innovation gets funded, over-investing in consumer apps and financial engineering while under-investing in harder, slower, less scalable problems (some areas of climate, health, and deep science). The fund-returner test (Module 10) that makes venture work is also, on this view, what narrows its social usefulness.
A third critique focuses on the misalignments this track has noted throughout: the fee-vs-carry tension (Module 09) incentivizing asset-gathering over returns; the markup games (Module 09's TVPI-vs-DPI) that let GPs show paper performance while raising the next fund; the founder secondary liquidity (Module 11) that can dull the alignment the model depends on. On this view, the venture model's incentives have drifted away from the simple "everyone wins only at a great exit" alignment that made it work, toward a set of games that benefit insiders.
These critiques are serious and partly true — and also partly answered by venture's defenders. The superabundance concern is real but may be cyclical rather than permanent. The "funds society poorly" critique is true on its own terms but venture has also funded genuinely transformative companies (the ones this track has used as examples). The "model is gamed" critique identifies real misalignments but the best firms still operate with strong alignment. The mature position is not to dismiss these critiques or to accept them wholesale, but to hold them as live tensions that the next decade will test. A student who can articulate the strongest case against venture understands the industry better than one who only knows the promotional version.
Amid all the forces and uncertainties, it's worth ending on what is unlikely to change — the structural features of venture finance that have persisted through every prior wave and will probably persist through the AI wave too. These are the load-bearing ideas of the whole track, and they're durable precisely because they reflect deep facts about funding high-uncertainty innovation.
The power-law distribution of outcomes (Modules 01, 04, 10) is not a feature of any particular era — it reflects the fundamental nature of funding high-uncertainty innovation, where a few bets succeed enormously and most fail. No change in fund structure, geography, or technology removes this. As long as venture funds genuinely uncertain new things, the returns will concentrate in a handful of winners, and everything else (portfolio construction, the fund-returner test, the need for ownership) follows from it.
Despite AI tools and quantitative methods, the core venture act — identifying a promising founder and an uncertain opportunity early, and choosing to back it — remains a matter of judgment under deep uncertainty. The information needed to make these decisions well is qualitative, relational, and forward-looking in ways that resist full automation. The best investors have consistently been distinguished by judgment, not by access to better data. This is likely durable even as AI augments the analytical work around the decision.
At its heart, venture is a relationship between people who build companies and people who fund them, negotiated under uncertainty and sustained over years (Modules 03, 04). The instruments change, the structures evolve, the geography shifts — but the fundamental transaction is two parties betting together on an uncertain future, with aligned-but-not-identical interests. This human core has been remarkably stable across every era of venture, and there's little reason to expect it to disappear.
Venture finance in 2026 is being reshaped by powerful forces — AI, the liquidity revolution, capital superabundance, geographic rebalancing, new structures — and faces serious critiques about its size, its social allocation, and its incentives. Much about its future is genuinely uncertain. But the deep structure endures: the power law that concentrates returns, the primacy of human judgment under uncertainty, and the founder-investor relationship at the core. A student who understands both the changing forces and the durable structure can reason about venture's future without being captured by either breathless optimism or reflexive skepticism. That balanced, structural understanding — built across this entire track — is what lets you read venture clearly as it evolves.
Six questions on the forces reshaping venture, the open questions they raise, and what remains structurally durable.