Moonshot capitalism: AI rewrites the venture capital playbook
Author: Gammatek ISPL, Published: Sep 2026 If you're evaluating enterprise software right now, whether that's an ERP system, a SaaS platform, or an AI-native tool your team wants to pilot, you're buying into a venture capital market that looks nothing like it did three years ago, and that has direct consequences for whether your vendor will still exist, still be funded, or still be independent in eighteen months. AI companies absorbed roughly 86% of all US venture capital dollars in the first half of 2026, according to PitchBook-NVCA data, up from 40-45% earlier in the same year. That's not a story about Silicon Valley excess. It's a story about where the next round of funding for the software you're about to depend on is, or isn't, going to come from.
In this article
The numbers behind "moonshot capitalism"
Why VC stopped looking like venture capital and started looking like infrastructure investing
What this means for enterprise SaaS specifically
The valuation gap, scored: AI-native vs. traditional enterprise SaaS
Vendor stability considerations before you sign a contract
Implementation checklist for buyers navigating this market
FAQ
The numbers behind "moonshot capitalism"
The scale of the shift is easier to grasp as a sequence than as a single statistic. In 2023, AI startups captured roughly 25% of global venture funding. By 2024, that reached 38%. In the first half of 2025, it hit 53% globally, with US concentration already at 64%. By Q1 2026, Crunchbase put AI's share of global venture funding at 80%, and PitchBook-NVCA data for the first half of 2026 showed AI companies had absorbed $355.9 billion of the $412.7 billion in total US venture capital deployed, close to 86% of every dollar. Global H1 2026 venture funding hit a record $510 billion, already closing in on the entire 2021 full-year peak of $643 billion, achieved in six months instead of twelve.
The concentration inside that number is the part that should actually change how you think about vendor risk. OpenAI and Anthropic alone accounted for 43% of all H1 2026 US venture funding. In Q1 2026 specifically, four companies, OpenAI, Anthropic, xAI, and Waymo, absorbed a combined $188 billion, roughly 65% of all global venture investment in a single quarter. Rounds of $100 million or more made up 87.5% of all capital deployed in that period, up from a sub-$100 million share of 43.8% just two years earlier in 2024. This is the "moonshot capitalism" the topic name refers to: venture capital increasingly behaving like sovereign infrastructure investment in a handful of frontier bets, rather than a broad-based financing market for thousands of early-stage companies across every category.
Why VC stopped looking like venture capital and started looking like infrastructure investing
Traditional venture capital was built on a portfolio logic: fund many companies, expect most to fail, let a small number of outsized winners return the fund. That model assumed capital was the scarce resource and ideas were relatively abundant, so spreading bets across many founders made sense. What's happening in 2026 inverts that assumption for a specific slice of the market. Frontier AI labs require capital at a scale, tens of billions of dollars per round, that only a small number of investors can supply, and the strategic value of controlling or having privileged access to a leading model has become valuable enough that late-stage investors are willing to pay for it directly, rather than betting on a portfolio of smaller companies that might eventually get there.
One venture partner described the resulting capital structure to TechCrunch as looking "more like the oil industry in the early 20th century than traditional venture capital," a comparison to strategic resource control rather than diversified financial betting. The scale comparison to the last era of extreme sector concentration makes the shift concrete: Q1 2026's roughly $300 billion in global venture funding is close to ten times the peak quarterly venture total during the dot-com boom, roughly $28 billion in Q2 2000. But where the dot-com era spread that capital across thousands of smaller bets on companies with wildly varying business models, 2026's record is built on a handful of unprecedented wagers at the very top. AI's 80% share of a single quarter's venture funding is the highest concentration of capital in a single technology sector in venture capital history, surpassing even the dot-com era's peak concentration in internet companies.
The practical effect for everyone outside the top four or five companies is a genuine funding drought that coexists with headline-grabbing total dollar figures. Early-stage funding rose a comparatively modest 41% year-over-year in Q1 2026, while late-stage funding rose 205% in the same period. The Series A crunch for non-AI startups that started in the ZIRP unwind of 2023-2024 hasn't resolved, it's been overshadowed by a much larger number one layer up.
What this means for enterprise SaaS specifically
Enterprise SaaS sits in an unusual position in this new structure, it's neither a frontier AI lab commanding mega-rounds, nor purely a legacy category getting starved out. It's bifurcating internally, and the line runs through whether AI is genuinely structural to the product or bolted on as a feature.
Public enterprise SaaS valuations compressed hard through the same period AI funding exploded. The median enterprise-value-to-trailing-revenue multiple across public enterprise SaaS companies tracked by PitchBook fell to 3.3x at the end of Q1 2026, down from 4.9x at year-end 2025 and 6.2x at year-end 2024, a two-year decline of nearly half. At the same time, Series D and later AI companies traded at 6.6 times the valuation of their non-AI venture peers during H1 2026. That's not a modest premium, it's a structural repricing of what counts as a fundable enterprise software company.
The practical read for anyone selecting or renewing enterprise SaaS vendors: the traditional SaaS companies you've relied on for years are now competing for both customers and capital against a category that investors are willing to fund at multiples 6-10x higher, which changes their incentive to add real AI capability versus a marketing veneer of it, and changes which of your current vendors are likely to still be independently funded, versus acquired, down-rounded, or quietly wound down, in the next 18 months.
The valuation gap, scored: AI-native vs. traditional enterprise SaaS
This is the comparison that should actually inform a vendor risk assessment, not just a market-trends slide.
Company type | Typical 2026 revenue multiple | What's driving it | What it means for a buyer |
Foundation model labs (OpenAI, Anthropic-tier) | ~37.5x average, some 35-45x | Winner-take-most infrastructure dynamics; strategic capital creating a valuation floor independent of DCF math | Not your typical vendor category, but the capital they absorb is capital not available to everyone below them |
AI-native enterprise SaaS (AI is the core product) | ~15-30x ARR on growth rounds | Investors treating AI-native workflow ownership as a strategic asset, similar logic to the labs one tier down | Well-funded, but valuations this detached from revenue carry real correction risk if growth slows |
Traditional SaaS with genuine AI integration | Modest premium (roughly 1-3x multiple lift) over legacy peers, per SEG Research | Premium only applies when AI demonstrably improves retention and expansion, not when it's a feature checkbox | The safer long-term bet if the AI integration is real, but harder to verify from a sales deck alone |
Legacy SaaS, minimal or cosmetic AI | 3-7x ARR, compressing further | Buyers and investors increasingly filter out AI-in-name-only positioning | Highest near-term risk of funding gaps, acquisition, or forced consolidation |
One data point worth sitting with before trusting any vendor's AI claims at face value: a 2026 SEG Research and McKinsey analysis found that roughly two-thirds of buyers evaluating companies for the AI valuation premium saw only limited real AI adoption in the businesses they assessed, and separately, McKinsey found only about 6% of companies claiming AI-driven value had actually generated measurable AI-driven results. The premium is real for the minority who've earned it. It's being claimed far more broadly than it's actually being delivered.
What happens at the exit: M&A, not IPOs, is absorbing the squeeze
The funding side of this story gets most of the attention, but the exit environment matters just as much for anyone assessing vendor stability, because it determines what happens to a company that can't keep raising at AI-premium multiples. Through 2026, the consensus among M&A advisors is that select IPOs will return for profitable, scaled companies, but most exits for the enterprise SaaS companies squeezed by this new capital structure will be secondary buyouts or dual-track processes rather than public listings. That matters because an acquisition changes a vendor relationship in ways an IPO generally doesn't: pricing structures get renegotiated, overlapping products get sunset, and support teams get absorbed or eliminated on the acquirer's timeline, not yours.
There's also a geographic dimension worth knowing if your vendor stack includes non-US software. North American SaaS trades at approximately 4.8x median revenue versus 3.9x in Europe and a wider 3.0-6.0x range across Asia-Pacific. That valuation gap is actively driving cross-border acquisition activity, with better-capitalized US buyers targeting European enterprise SaaS assets specifically because they're priced at a discount to domestic equivalents. If your organization relies on a European or Asia-Pacific enterprise software vendor, that pricing gap makes an acquisition by a US-based, AI-premium-funded competitor a meaningfully higher-probability outcome over the next 18-24 months than it would have been in a less concentrated funding environment.
The pattern we've seen play out across enterprise software engagements is consistent: the vendors most exposed to a forced sale aren't the ones with the weakest product, they're the ones with solid, unglamorous, mid-single-digit-ARR-multiple businesses that can't credibly claim the AI premium and can't compete for talent or marketing spend against AI-native rivals raising at 5-10x their valuation. That's a survivable position for the vendor's investors. It's a genuine planning risk for a customer who assumed "stable, profitable, and been around for a decade" was itself a form of vendor safety in 2026's market. It increasingly isn't, on its own.
The funding data above isn't abstract market commentary if you're the one signing a multi-year enterprise SaaS or ERP contract. A vendor's position in this reshaped capital market is now a legitimate part of procurement due diligence, alongside security posture and SLA terms.
Ask directly when the vendor last raised, at what multiple relative to ARR, and from whom. A company that raised at 3-4x ARR from a traditional SaaS-focused fund in 2024 is in a structurally different position than one that raised at 20x+ from an AI-focused growth fund in 2026, and that difference predicts runway, not just prestige.
Distinguish AI-native architecture from AI-feature bolt-ons in the product itself, not the pitch. Ask what specifically changes in the underlying product if the AI layer were removed. If the answer is "not much," you're likely looking at a legacy SaaS company applying AI-premium positioning it hasn't structurally earned.
Check whether the vendor's category is one investors are actively funding right now, or one being quietly starved. Application-layer AI tools built on top of foundation models received comparatively modest funding relative to the infrastructure and foundation-model tiers in 2026, which matters if your vendor sits in that squeezed middle layer.
Weight acquisition risk, not just shutdown risk. In a market this concentrated, the more common outcome for an underfunded but viable enterprise SaaS company isn't collapse, it's acquisition by a better-funded AI-native competitor, which can mean sudden pricing changes, product discontinuation, or a forced migration on someone else's timeline.
Implementation checklist for buyers navigating this market
Add a funding-stability question to your vendor evaluation scorecard. Most procurement checklists cover security and support SLAs but say nothing about capital position, which has become a materially larger risk factor in 2026 than it was three years ago.
Prioritize vendors with measurable AI-driven outcomes over vendors with AI-heavy marketing. Ask for retention and expansion data tied specifically to AI features, not aggregate company metrics that could be explained by other factors.
Build contract flexibility for the possibility of an acquisition. Negotiate data portability and reasonable exit terms now, while you have leverage, rather than during a forced migration after a vendor is acquired.
Reassess your vendor's category risk annually, not just at renewal. A company that looked stable at your last renewal may be sitting in a funding category, like the application layer squeezed between foundation models and infrastructure, that's losing investor appetite faster than its product is losing customers.
Don't assume "well-funded" means "stable." Several of the highest multiples in this market are running well ahead of the revenue growth needed to sustain them; a company that raised at 25x ARR carries real correction risk if growth decelerates, which is its own category of vendor risk distinct from being underfunded.
Weight non-US vendors for cross-border acquisition exposure specifically. If a vendor is priced at a discount to its US equivalents on a revenue-multiple basis, that discount is precisely what makes it an acquisition target; ask directly whether the vendor has fielded acquisition interest, and what commitments, if any, exist around product continuity if ownership changes.
Get support and roadmap commitments in writing, not in a sales deck. Verbal reassurance about a vendor's "long-term independence" carries no weight in an acquisition; a written commitment to a minimum support window and migration assistance does, and costs a vendor little to offer if they genuinely intend to honor it.
This is the same discipline we've walked clients through when evaluating AI governance and accountability gaps in enterprise tools and when assessing what the trust data means for AI vendors specifically: the fix isn't avoiding AI-adjacent vendors, it's building a real evaluation process instead of taking funding headlines or AI marketing claims at face value.
Frequently asked questions
Does this level of AI funding concentration mean a bubble is forming? The data supports both a real structural shift and genuine bubble risk simultaneously. The winner-take-most dynamics around foundation models reflect real infrastructure economics, but the gap between claimed AI value and delivered AI value, roughly two-thirds of evaluated companies showing limited real adoption despite AI-premium positioning, is exactly the kind of divergence that precedes a correction in the most overextended part of the market, even if the frontier labs themselves remain well-capitalized.
Should we avoid vendors that haven't raised an AI-premium round? No. A vendor with modest, sustainable multiples and real revenue is often a safer long-term partner than one carrying a 20-30x ARR valuation it needs explosive growth to justify. The funding data is a risk input, not a simple better-or-worse signal.
How is enterprise SaaS different from AI-native SaaS in practice? Enterprise SaaS traditionally sells workflow software with AI as an added feature layered onto an existing product. AI-native SaaS is built with AI as the core mechanism of the product itself, meaning the product's value proposition doesn't really exist without it. Investors are pricing that distinction explicitly now, with AI-native platforms commanding 25-30x EV/Revenue versus roughly 3-4x for traditional public SaaS.
What's the single most useful question to ask a vendor about their funding position? "What multiple of ARR did your last round value you at, and who led it?" The answer tells you more about the vendor's runway, growth expectations, and pressure to show rapid returns than almost any other single data point you can get in a sales conversation.
Is this concentration specific to the US, or global? It's most extreme in the US, where AI companies captured roughly 86% of venture dollars in H1 2026 versus roughly 70% globally, but the trend direction is the same everywhere: AI captured just 15% of global venture funding in 2020, and the share has climbed every year since.
If most exits are acquisitions rather than IPOs, how would we even know our vendor is being acquired before it happens? You often won't get meaningful advance notice, which is exactly why the contract terms matter more than the warning signs. Data portability clauses, defined minimum support windows, and change-of-control provisions negotiated before a deal is on the table are the practical protection, since public rumor or leaked term sheets are not a reliable early-warning system and by the time either surfaces, negotiating leverage has usually already shifted to the acquirer.
We're a mid-sized company. Does any of this actually apply to us, or is it only relevant for enterprise-scale buyers?It applies at any size that depends on a vendor's continuity. Mid-market buyers are often more exposed than large enterprises to a sudden vendor acquisition or shutdown, since smaller accounts typically have less negotiating leverage to demand migration assistance after the fact and less internal capacity to absorb an unplanned platform switch on short notice.
Not sure whether your current enterprise software vendors are funded well enough, or in the right category, to still be around at your next renewal?
Our enterprise software and vendor risk assessment reviews your current SaaS and ERP stack against funding stability, AI-claim verification, and contract exposure, so a vendor's funding headline doesn't become your migration emergency.
Comments