96% of private equity investors have already changed how they value software companies. The same shift is hitting crypto—but nobody is talking about it.
A recent survey by Lazard, a global investment bank, dropped a quiet bomb. They polled institutional investors in the private equity secondary market. The question: How has AI changed your view on software companies? The answer: Almost everyone has moved. 96% have altered their investment approach. 91% now say that “proprietary data advantages and network effects” are the only real moats. And 4%—barely a rounding error—haven’t changed a thing.
This isn’t just a Wall Street story. It’s a crypto story. Because the software that powers crypto—dApps, DeFi protocols, Layer 2 sequencers, wallet infrastructure, even the code that runs your favorite NFT marketplace—is facing the exact same AI-driven reckoning. The difference? Crypto software is built on public data. Transparency is a feature, not a bug. But that very transparency makes the “data moat” argument a ticking time bomb.
I’ve been in this game since 2017. Back then, I built a Python script to scrape 150 ICO whitepapers and break down tokenomics overnight. I learned one thing: speed is the only edge that matters. But speed without data is noise. The same applies today. The Lazard survey tells me that the smart money is already moving. They’re not waiting for the chart to scream. They’re front-running the panic.
The chart whispers before the market screams.
Let’s decode the core signal. The survey’s 96% behavior change isn’t abstract. It means capital is flowing out of traditional software assets and into… something else. The report says “investors are moving funds to other opportunities.” My read: they’re rotating into AI-native infrastructure, or they’re demanding higher risk premiums from software assets that can’t prove an AI moat. In crypto, we see the same pattern. Look at the valuations of DeFi protocols that rely on front-end monopoly. A pure copy-paste of Uniswap’s interface? No moat. But a protocol that owns unique on-chain data pipelines—like a DEX that aggregates liquidity from 10 chains and uses AI to optimize routing? That’s a different story.
The core insight: 91% of investors believe data moats + network effects are the only sustainable defenses. In crypto, that’s dangerous.
Public blockchains are the ultimate data commons. Every transaction, every swap, every liquidation is visible. An AI model can ingest all of it. The so-called “proprietary data” of a crypto protocol is, by design, open for anyone to scrape. The only real moat in crypto is not data ownership—it’s composability. The deep integration of a protocol into a user’s workflow. The liquidity that can’t be easily moved. The trust that comes from years of uptime. That’s what the Lazard survey misses.
Contrarian angle: The consensus is the trap.
When 91% of investors agree on a single moat, that moat is already priced in. The real alpha lies in what they’re ignoring. For crypto software, the overlooked moat is liquidity depth and composability stickiness. An AI agent can analyze order books, but it can’t recreate the trust that comes from 500,000 users who have never been rugged. The Ethereum network effect isn’t about data—it’s about the collective willingness to pay for blockspace. That’s a social moat, not a data moat.
Liquidity is the only truth that bleeds.
And here’s the kicker: The survey’s 4% who didn’t change? They might be the smartest ones. They’re the ones who realize that AI is a feature, not a business model. The crypto software companies that win will be those that embed AI as a tool to enhance user experience, not as a replacement for their core value proposition. Think of a Layer 2 that uses AI to predict congestion and adjust fees dynamically. Or a wallet that uses AI to detect phishing attacks before the user clicks. That’s augmentation, not substitution.
Speed is the new currency of trust.
We trade the panic, not the price. The panic right now is that AI will commoditize every software product. But in crypto, the product is often the network itself. You can’t commoditize a distributed ledger. You can’t AI-away the security of a PoS validator. The real risk is for centralized crypto software providers—like hosted wallets, centralized exchange interfaces, or custody dashboards. Those are the ones facing the same 96% squeeze.
Takeaway: Watch for three signals over the next 12 months.
First, track the trading volume of crypto software shares on secondary platforms like Forge Global. If they start trading at a discount to NAV, that’s the market pricing in AI disruption. Second, monitor the AI feature adoption of top DeFi protocols. If Uniswap integrates an AI-powered routing engine that cuts slippage by 30%, that’s a moat. Third, look at the funding rounds of AI-native crypto startups. The cheetah runs when the gazelle is tired. The next cycle’s winners will be the ones who decode chaos into order—not the ones who hoard data that anyone can see.
The code is cold, but the hype is hot.
I’ve been through the 2017 ICO rush, the 2020 DeFi summer, the 2021 NFT mania, and the 2022 collapse. Each time, the signal was buried in the noise. This time, the signal is clear: AI is redefining the value of software, and crypto software is no exception. But the playbook is different. Don’t chase the 91% consensus. Chase the 4% that are quietly building the next generation of AI-native, composable, trust-minimized protocols. That’s where the real alpha hides.
