Goldman Sachs CEO David Solomon made a striking statement at an a16z founder event:
"I think this could be the biggest M&A year in history. Fiscal stimulus, monetary stimulus, a rate cutting cycle… this cocktail of stimulus is very hard to slow down."
That single statement has every AI founder paying close attention — and for good reason. Global M&A deal value hit a record $4.9 trillion in 2025, with nearly half tied to AI transactions. In Q1 2026 alone, 266 AI M&A deals closed — a 90% year-over-year increase. And this wave hasn't crested yet.
The question "should I sell now?" has never felt more strategic.
- Why Goldman's CEO is calling 2026 the biggest M&A year ever
- Why big companies are buying AI startups instead of building
- How much AI-native companies are fetching in 2026
- How founders should judge the right moment to sell
- 3 practical actions to prepare for an acquisition
What's Actually Happening?
An M&A supercycle isn't just "a lot of deals happening." It's a structural inflection point where capital, strategic demand, and regulatory conditions shift simultaneously from buyer-favored to seller-favored markets.
a16z partner David Haber summarized four key drivers behind the current moment:
- Throwing money at AI actually works
Traditional software followed the mythical man-month principle — more people or dollars didn't accelerate outcomes. AI breaks that rule. More data and compute (= more capital) genuinely improves results. Capital has become a competitive edge, which is why large companies are betting aggressively on AI acquisitions. - Global M&A is recovering strongly for the first time since 2022
Pent-up M&A demand — suppressed by rate hikes and tightening regulation from 2022 to 2024 — began recovering in Q4 2025. The BCG M&A Sentiment Index is rising across financial services, tech, and healthcare. - $100M+ AI deals are at a record share of VC capital
In 2025, the share of VC capital flowing into $100M+ AI deals reached historic levels — a clear signal of capital concentration in AI. - The four hyperscalers invested $350B+ in AI capex in 2025 alone
Amazon, Google, Microsoft, and Meta's infrastructure spend directly creates demand for the software and applications that run on top. Buy the infrastructure, and you need to buy the products that use it.
What's Different This Time?
Looking at why large companies are buying AI startups — and at what price — makes it clear this cycle is structurally different from 2021.
| Factor | 2021 M&A Cycle | 2026 AI M&A Cycle |
|---|---|---|
| Acquisition motive | User base, market share | AI capabilities, data, talent package |
| Acquirer pool | Big Tech only | Big Tech + Enterprise SaaS + PE |
| Valuation driver | Revenue multiples | Data moat + NRR 120%+ |
| Traditional SaaS multiple | 8–12x ARR | 4–6x ARR (declining) |
| AI-native multiple | — | 8–15x ARR (up to 35x+) |
| Deal structure | Traditional 100% acquisition | Acqui-hire, licensing hybrids |
The actual deals tell the story: Alphabet acquired Wiz for $32B, ServiceNow bought Moveworks for $3B, Workday closed Sana for $1.1B. But the structural shift is even more telling. Meta's $14.3B investment in Scale AI plus CEO hire, Google's $2.7B licensing deal to bring back Character.ai researchers — hybrid deals that achieve acquisition outcomes while navigating antitrust are becoming the new standard.
According to FE International's analysis, building AI capabilities from scratch in 2026 costs more and takes significantly longer than acquiring them. Enterprise buyers have officially crossed the threshold from "build" to "buy." That's why the acquirer pool has expanded from Big Tech to enterprise SaaS and PE firms.
What Founders Should Do Now
A supercycle is an opportunity — but only for those who are prepared. Here's a checklist based on what separates premium AI exits from average ones in 2026.
- Prove your AI drives measurable business outcomes
Acquirers are no longer paying premiums for "we use AI." They want to see AI functionality driving NRR above 120%. If your AI features aren't moving business metrics, the premium disappears — no matter how impressive the demo. - Document your data moat
Proprietary data is the most durable competitive advantage in AI. Clearly articulate why your data set is non-replicable: data provenance, governance practices, and how the model improves with usage. This is your strongest card at the deal table. - Prepare financials with full AI cost loading
Acquirers will scrutinize inference costs, model dependencies, and compute commitments. Know your gross margins with AI costs fully loaded, and how your cost structure evolves at scale. Clean answers here compress deal timelines and protect valuation.
1. Do you own proprietary data and workflows in a specific vertical?
2. Is your NRR above 120% or trending that direction?
3. Can you prove "our customers changed because of AI" — not just "we use AI"?
If all three are yes, now may be the optimal moment to start an M&A process.
One more thing to keep in mind: supercycles always cool. Capital cycles shift, acquirer priorities evolve, regulatory environments change. As FE International puts it: "Waiting for a better time is a strategy that works until it doesn't."
