Venture Capital Enters Its Third Act
May 2025
The Age of Intelligence
If each era of technology is defined by its dominant metaphor—the mainframe, the personal computer, the cloud—the current one may well be symbolised by the language model. Artificial intelligence, once confined to research labs and sci-fi scripts, is now steering the course of industry itself. Much as the web did in the 1990s and mobile computing in the 2000s, AI has inaugurated a new technological cycle. And venture capital, its most fervent patron, is evolving alongside it.
Startups no longer pitch AI as window dressing. They build with it at the core. From legal advice (Harvey) to healthcare research (OpenEvidence) and customer relations (Day.ai), a new species of AI-native firms is taking shape. Meanwhile, incumbents like Notion and Attentive are layering generative tools atop legacy software, turning 10x productivity promises into real-world margins. The distinction between “AI-native” and “AI-enhanced” is quickly becoming academic.
For founders, AI lowers the marginal cost of experimentation; for investors, it sharpens the speed and scope of returns. Operational processes—from engineering to customer support—are being redefined, if not outright reinvented. The toolkit of company-building is lighter, faster, and more formidable than ever.
Three Epochs of Venture Capital
The venture industry, like the technology it funds, tends to reinvent itself every generation. Its present transformation—driven by AI, global capital, and sectoral convergence—marks the beginning of what might be called VC 3.0.
VC 1.0: Cottage bets and silicon dreams (1948–1994)
In its earliest decades, venture capital was an elite and intimate affair. Fewer than 200 general partners roamed Sand Hill Road, backing semiconductor pioneers with modest sums and enormous equity stakes. A typical investment might involve $2m in exchange for 40% of a firm—after the founder had bootstrapped it into double-digit millions in revenue. Legends such as Don Valentine and Arthur Rock operated with the discretion of private bankers and the instincts of gamblers. Publicity was scarce; capital scarcer still.
VC 2.0: The rise of software and spectacle (1994–2023)
The commercialisation of the web in the mid-1990s flung open the doors. With browsers came broadband, with broadband came billions. Software’s seductive economics—low fixed costs, infinite scalability, global reach—reshaped both venture economics and cultural imagination. Accelerators emerged; term sheets became standardised; founders were profiled in glossy magazines before they turned a profit. The number of active VCs exploded into the tens of thousands. VC business went international!! A whole “startup industrial complex” was born, spanning media, education, and professional services. The names—Sequoia, Benchmark, Andreessen Horowitz—became brands in their own right.
VC 3.0: Capital meets cognition (2023– )
Now, the game is changing again. With generative AI as its engine, venture capital is beginning to disrupt itself. AI is not merely what firms invest in—it is how they invest. Deal sourcing, due diligence, and portfolio support are all being augmented or automated. Human intuition is being flanked, if not replaced, by probabilistic scoring, pattern recognition, and large language models.
The ambit of venture capital is widening, too. It is no longer limited to pure software plays or mobile apps. Capital is flowing into defence, climate technology, synthetic biology, and space—sectors once dominated by government grants or corporate R&D. Software may have eaten the world, but AI is now chewing through its most intractable domains.
Venture capital’s next phase will be larger, more global—and less intimate ?
In 1990, venture capital was a niche corner of finance. Today, it is a $200bn-a-year industry with over 32,000 active practitioners and a footprint that stretches from Silicon Valley to São Paulo, from Bangalore to Berlin. And yet, by historical standards, it remains small. For all the fanfare, venture capital still accounts for less than 1% of global investable assets. That, too, is set to change.
The conditions for growth are plainly in place. In a world starved of alpha, venture capital continues to deliver outperformance—albeit unevenly. The market remains inefficient, its returns driven as much by insight and access as by structure or scale. For institutional investors with long horizons and an appetite for risk, few other asset classes offer a similar mix of returns, narrative, and optionality.
The market itself is expanding. What was once confined to software is now sprawling across deep tech, defense, climate, and the biosciences. Governments—particularly in emerging economies—have discovered that venture capital is not merely a source of growth, but a lever of policy. From Riyadh to Jakarta, it is being mobilised as an instrument of economic diversification, job creation, and geopolitical soft power.
This expansion is also changing the character of the industry. LP appetite is fragmenting into niche strategies—from climate funds to AI-specific mandates—while the LP base itself is broadening. Sovereign wealth funds, family offices, and corporate VCs are all muscling in. At the same time, better tools and data have lowered the barriers to entry. Angel syndicates, rolling funds, and operator-led vehicles are drawing in thousands of part-time capital allocators, blurring the line between professional investor and enthusiastic amateur.
Yet as the industry scales, a more subtle transformation is taking place: a shift in culture. For all its gloss, venture capital has long prided itself on its missionary zeal—its openness to outsiders, its reverence for founders, its appetite for risk. This ethos is now under quiet strain.
The influx of capital—and the gravitational pull of larger deals—has pushed much of the industry into later-stage territory, where returns are more predictable, due diligence more rigorous, and terms more aggressive. In this zone, venture capital starts to resemble private equity. The investor’s role moves from nurturer to negotiator. Control matters more than conviction.
This convergence is not necessarily malign. There is merit in discipline, and some founders may benefit from sharper governance. But the shift has cultural consequences. Where venture capital once championed bold bets on improbable ideas, it now flirts with metrics, margins, and mezzanine debt. The romance of the garage startup gives way to the calculus of structured growth.
The danger is not overreach, but drift. Venture capital risks becoming just another flavour of finance: professionalised, institutionalised, and risk-averse. That would be a loss—not just for founders, but for the economies and societies that depend on genuine innovation.
For the industry to remain vibrant, it must preserve the spirit that made it formidable in the first place: an openness to uncertainty, a belief in talent before traction, and a willingness to back vision before product.
The Evolving Landscape of Venture Capital: Convexity, Constraints, and a Coming Reset
For decades, venture capital has operated on a logic few other asset classes would dare emulate: most bets fail, a handful succeed, and one in thirty might just pay for everything. The model is not built for balance—it is built for asymmetry. Convexity, rather than consistency, is the game.
A typical fund might back 20 to 30 companies, knowing that the vast majority will disappoint or disappear. But one “fund-returner”—a Google, a Shopify, an Airbnb—can make up for all the rest. This winner-take-most dynamic has created fortunes, legends, and an industry mythology built on the power of the outlier.
Yet for all its internal coherence, the structure is stubbornly exclusionary. Retail investors, for instance, have little hope of replicating such a strategy. Without the capital, deal access, or rights to participate in follow-on rounds—where most of the value accrues—they are left at the margins. Even early exposure offers little reward if follow-up funding is closed to outsiders. The result is a market that remains resolutely closed, structurally tilted toward insiders, and largely indifferent to those beyond its gates.
There is now little doubt that venture capital has outgrown its origins. Once a high-risk, relationship-driven cottage industry, it has ballooned into a global asset class with geopolitical relevance and cultural cachet. The largest firms have become platforms: multi-stage, multi-product, multi-geography investment machines. In ambition and structure, they resemble private-equity titans more than their scrappy predecessors.
Scale brings advantages—greater access to capital, stronger brand equity, the ability to support companies across stages and cycles. But it also exacts a cost. Decision-making slows. Attention thins. Incentives drift from performance to asset gathering. The spirit of conviction—once the hallmark of the industry—can give way to passive indexing, disguised as diversification.
In retrospect, the go-go years of 2020 and 2021 appear less a golden age than a distortion. Capital was abundant, diligence cursory, and valuations euphoric. Megafunds such as Tiger Global and SoftBank sprayed capital with speed and abandon, creating an illusion of perpetual growth. That illusion has since evaporated.
As rates rise and liquidity tightens, the effective cost of capital—regardless of central bank pronouncements—has soared. Many funds now find themselves under-reserved, caught off guard by the disappearance of easy follow-on financing. The herd is being culled. Those who scaled too fast, sacrificed focus, or treated venture as an allocation exercise rather than a judgment call are fading from view.
Lean teams with focused theses and high barometers for quality are outperforming. Constraint, long seen as a limitation, is being reframed as a competitive advantage. Smaller partnerships, closer to the founders they back, are rediscovering an old truth: in a business that bets on the improbable, discipline is not a luxury—it is a necessity.
The Shape of Things to Come
As venture capital grows up, it is also growing sideways. The language of late-stage investing is increasingly indistinguishable from that of private equity: consolidation, operational roll-ups, financial engineering. Terms like “margin expansion” and “capital stack optimisation” now circulate in pitch decks once devoted to product-market fit. Add a layer of AI-enhanced diligence, and the pitch becomes sharper still—if not necessarily more sound.
Yet for all the sophistication, there is unease. PE-style tactics, when applied to early-stage companies, often miss the point. Venture capital is about embracing volatility, not suppressing it; about backing vision, not merely optimising EBITDA. A strategy designed for stable cash flows and mature markets may look elegant on a spreadsheet. It can prove disastrous in a sandbox.
Still, the venture model is far from broken. If anything, its messiness remains a feature, not a flaw. The industry is riddled with inefficiencies: opaque pricing, fragmented access, and inconsistent underwriting standards. For those with judgment—and luck—these conditions continue to offer outsized rewards. That promise, coupled with structural tailwinds, suggests that the market may well expand from $200bn annually to something far larger.
Much of that growth will come from geography. In Latin America, the Middle East, Southeast Asia, and Africa, governments are treating startups not as curiosities but as industrial policy. Sovereign funds and development agencies are allocating capital. Domestic venture ecosystems are maturing. Mobile infrastructure and digital rails—once obstacles—are now accelerants.
Within developed markets, too, the cast of investors is changing. Where once Sand Hill Road reigned supreme, new entrants are gaining ground: corporate arms hungry for growth, family offices in search of yield, and a rising class of solo capitalists. Enabled by syndicates, platforms, and AI-driven tools, these micro-allocators can now manage portfolios that previously required institutional scale.
Perhaps the most intriguing challenge to venture orthodoxy is coming from below. A wave of new models aims to loosen the gatekeeping. One such idea, “venture staking,” offers retail investors the right to participate in future funding rounds—effectively an option on access. Proponents see it as a way to unlock trillions in sidelined capital and widen the circle of participation.
This, too, reflects a shift in what it means to be venture-backable. The old bar—exponential growth, polished decks, and elite credentials—is giving way to a broader aperture. New technologies enable capital-light models. New founders come from overlooked geographies and non-traditional backgrounds. The startup has become a format as much as a company: a flexible shell for experimentation, not just an engine for hypergrowth.
As venture capital expands—geographically, structurally, and ideologically—it risks losing the clarity that once defined it. A business built on intuition and concentrated bets is flirting with models of scale, standardisation, and financial abstraction. That may be the cost of maturity. Or it may be a signal that venture capital is once again due for reinvention.
The challenge, as ever, will be to preserve the core ethos—risk, vision, and long-termism—while accommodating a broader, noisier, and more democratic market. If venture capital is to remain the engine of innovation, it must resist becoming a mere variation on conventional finance. Its future lies not in mimicking private equity, but in staying just strange enough to fund the improbable.
In the heady world of startup finance, valuations are often more narrative than number. Nowhere is this more evident than in the club of unicorns—private companies supposedly worth $1bn or more. Yet as Professor Ilya Strebulaev of Stanford has shown, the headline figures are often misleading. Strip away the fine print—liquidation preferences, ratchets, participation rights—and the true value may be 30% or more below the sticker price.
Such discounts are not just academic curiosities. They reveal a growing fissure between perception and substance in late-stage venture. Inflated valuations fuel media cycles, boost egos, and attract capital. But they also inflate expectations, distort benchmarks, and complicate exits. What begins as marketing ends as mispricing.
This is more than a problem of optics. At the portfolio level, the influx of capital into overvalued companies crowds out discipline. Scarcity, once the venture investor’s greatest ally, is being replaced by abundance—of funds, of follow-ons, and of hype. In such conditions, selectivity falters. And when the music stops, it becomes painfully clear who was underwriting growth, and who was merely underwriting momentum.
Yet amid the froth, a more grounded model is quietly emerging. Firms are shrinking in size, sharpening their theses, and leaning more heavily on technology. The shift is partly economic—rising costs, slower exits, and the need to do more with less. But it is also strategic. In an industry defined by asymmetric outcomes, bloated teams and generic strategies are liabilities, not assets.
GPUs may replace interns; dashboards may edge out Monday-morning memos. But the underlying philosophy remains unchanged: back founders with outsize ambition, intervene at moments of scale, and exit with both timing and conviction. It is not a volume game. It is a judgment game.
Artificial intelligence, too, will reshape the firm—though perhaps not in the ways most imagine. AI will not supplant the venture capitalist’s gut. But it will complement it, surfacing patterns, filtering noise, and compressing decision cycles. The edge will belong to those who use machines to augment—not replace—their ability to form and act on convictions.
As venture capital ages, it faces a paradox. The temptation is to professionalise: to borrow the language of asset management, to scale operations, to abstract away from the founder. But the risk is that in doing so, it loses the very quality that made it effective—a capacity to believe early, to risk boldly, and to see past consensus.
The firms that thrive in the coming decade will not be the largest, the loudest, or the most diversified. They will be the most disciplined. In a market increasingly defined by noise and access, clarity of purpose may once again become venture’s rarest asset.
Certainly. Here's a refined and structured version of your article in the style of The Economist—sharp, balanced, globally aware, and confident without being breathless:
Rebuilding the Machine
Venture capital is being reinvented—again. As the industry matures amid technological upheaval, macroeconomic headwinds, and geopolitical realignment, its centre of gravity remains unchanged: the founder. Yet around this constant, almost everything else is shifting. AI, new funding models, regional ecosystems, and cultural frameworks are reshaping what it means to be a venture capitalist in the 2020s.
In a world awash with capital and ideas, the team remains the scarce resource. Talk of network effects and AI moats fills pitch decks, but seasoned investors know better. Execution trumps inspiration. As one puts it, “Ideas are overrated. It’s the team that matters.”
That has pushed venture capitalists further into the role of talent scouts. Pattern recognition, once a euphemism for bias, is being refined into a disciplined search for rare traits: resilience, obsession, and the ability to navigate ambiguity. The rise of “founder-forward” investment structures—such as income-sharing agreements and pooled equity vehicles—suggests a growing recognition that people, not products, are the core asset class. After all, the best entrepreneurs have a chip on their shoulder and something to prove.
Silicon Valley still looms large, but its monopoly on innovation is fading. The map of venture is redrawing itself. Tel Aviv, London, Singapore, Nairobi, and Bangalore are now permanent fixtures on the investment circuit. Sovereign wealth funds and local LPs are funnelling capital into homegrown talent. Governments, eager for jobs and soft power, are drafting innovation into their national strategies.
But globalisation cuts both ways. Rising geopolitical risk—from US-China decoupling to internet fragmentation—may curb cross-border deal flow and dent investor confidence. Some warn of a “re-balkanisation” of venture capital, as regulatory barriers rise and global syndicates become harder to assemble. The irony is rich: just as capital becomes more global, trust is becoming more local.
As markets wobble and retail appetite for innovation grows, venture is seeing a quiet renaissance in financial innovation. One model gaining traction is “venture staking,” which allows non-institutional investors to pre-purchase rights to join future rounds—potentially unlocking billions in untapped capital. Another trend is the proliferation of “founder guilds”: pooled-capital structures supporting individuals across multiple ventures, insulating serial entrepreneurs from the one-shot pressure of startup failure.
On the supply side, general partners are also experimenting with internal architecture. Some funds assign a “devil’s advocate” to each deal; others review their anti-portfolios—the companies they passed on that later thrived—as a form of institutional humility. Anti-veto provisions now empower lone dissenters to back contrarian bets, reinstating what venture was meant to reward: conviction over consensus.
In a post-boom world, leaner teams are performing better. Many of the large funds built during 2020–2021’s liquidity surge are struggling with bloated structures, diluted founder relationships, and internal misalignment. By contrast, smaller, thesis-driven firms—focused more on judgment than spreadsheets—are showing stronger returns.
But the constraints are not merely structural. They are personal. Many GPs, lured by management fees and institutional scale, face what one investor calls “the money trap”—where focus is sacrificed for fundraising. In a craft that rewards obsession, curiosity can be a double-edged sword: diverse interests may fuel imagination, but they can just as easily fragment attention.
Beneath the bravado, venture capital is a business of radical uncertainty. That is its burden and its edge. The best investors are those who operate not despite the unknowns—but because of them. Yet even they face a more demanding landscape. Legal frameworks for emerging models like income-sharing or staking remain untested. Regulatory scrutiny is rising. LPs, chastened by the correction, are more discerning—and more impatient.
Venture’s next chapter will not be written by the largest platforms or the deepest pockets. It will be shaped by firms that embrace a quieter approach: smaller, more disciplined, more global, and more grounded. The best of them will invest not just in companies—but in character, context, and contrarian insight.