Last week, Lazard released a survey of private equity secondary market investors that, on its surface, is about traditional software. But read it through the lens of macro liquidity flows and it becomes a distress signal for the entire technology stack—including crypto. The headline numbers are stark: 91% of respondents now view proprietary data plus network effects as the only defensible moat. Only 4% have not changed their investment methodology. The rest have already begun shifting capital to other opportunities.
This is not a mere opinion shift. It is a structural repricing of an entire asset class. And it is happening at the exact moment that crypto’s Layer2 ecosystem is fragmenting liquidity into a thousand pieces, each claiming to be the next scalability solution while competing for the same shrinking pool of users.
I have been watching this pattern for years. In 2017, I spent six months auditing Ethereum’s early DAO experiments, pouring €15,000 into a prototype that collapsed with the Parity wallet hack. The lesson was not about code—it was about the gap between theoretical decentralization and practical security. That gap is now widening in both software and crypto, as AI commoditizes the layer of functionality that once commanded premium valuations.
s chaotic surface — The Lazard survey reveals a quiet consensus: code is no longer a moat. AI can generate it. APIs can replicate it. The only thing that remains scarce is the data that feeds the models and the network that distributes the output. This is the same logic that makes Bitcoin’s security model fragile without Ordinals, and that makes Layer2s a liquidity-slicing exercise rather than true scaling.
Context
The survey, conducted by Lazard’s secondary advisory group, polled institutional investors active in the private equity secondary market. These are the players who buy and sell stakes in venture-backed software companies, providing liquidity to LPs and GPs. Their collective behavior is a leading indicator for the broader asset class.
The key findings: - 91% believe proprietary data and network effects will be the primary moat in an AI-dominated world. - Only 4% have not changed their investment approach. - A significant portion are moving capital to other sectors entirely.
Lazard’s report is dated August 2023? Or 2024? The exact year is less important than the signal: the market has already priced in a paradigm shift. The old valuation framework—based on MRR multiples, growth rates, and net dollar retention—is being abandoned. The new framework is still forming, but its contours are visible: AI exposure discount, data asset quality, and network density.
Core
This is where the crypto parallel becomes uncomfortable. The software industry is undergoing a value anchor migration from code to data and network. Crypto, which has spent the last decade obsessing over code (smart contracts, consensus mechanisms, gas optimizations), faces the same disruption. But with a twist: crypto’s “data” is on-chain, transparent, and often public. Its “network” is fragmented across dozens of Layer2s, each with its own user base, liquidity pool, and security assumptions.
Let me be precise. The Lazard survey tells us that investors are now asking: does this software company own a unique data asset that cannot be replicated by a foundation model? And does it have a network effect that compounds with each new user? If the answer to either is no, the company is effectively a commodity, and its valuation will compress toward zero.
Apply this to crypto. Which protocols have proprietary data? Bitcoin’s UTXO set is unique, but it’s public. Ethereum’s state is massive but also public. The only “proprietary” data in crypto is off-chain: order flow from centralized exchanges, user behavior in DeFi apps, transaction metadata from private mempools. These are the assets that could become moats. But they are precisely the assets that most crypto projects do not own—they are owned by intermediaries like Coinbase, Binance, and the infrastructure providers.
s architectural silence — The quiet truth is that most Layer2s are not building data moats. They are building cheap copies of the same EVM execution environment, competing on fee discounts and marketing grants. The result is not a scalable ecosystem but a fragmented one, where each chain’s value proposition is identical to the next. This is exactly the kind of “middle-layer collapse” that the Lazard survey predicts for traditional software.
Contrarian
But here is the contrarian angle: AI might actually strengthen the one asset that crypto has always taken for granted—trustless settlement. The Lazard survey’s emphasis on “proprietary data” assumes that data is a scarce resource. In crypto, data is abundant but integrity is scarce. The ability to verify transactions without a trusted third party is a network effect that no AI model can replicate. Bitcoin’s security model, while threatened by the end of block rewards, is sustained by the very thing AI cannot fake: a provable chain of work.
This is the decoupling thesis. While traditional software is being disrupted by AI’s ability to generate code and analyze data, crypto’s core value proposition—permissionless verification—becomes more, not less, valuable. The Lazard survey’s investors are moving capital away from software that lacks data moats. But they are not moving away from Bitcoin. In fact, the opposite is happening: institutional capital is flowing into Bitcoin ETFs precisely because it is uncorrelated with the AI-driven software cycle.
s cold burn — The irony is that the same AI that is destroying software margins is also the force that will force crypto to mature. Layer2s that cannot demonstrate a sustainable data or network moat will die. The survivors will be those that either (a) own a unique off-chain data set (like a decentralized identity graph or a private transaction network) or (b) integrate AI into their core protocol in a way that creates a feedback loop (like a DeFi lending market that uses AI to price risk dynamically).
Takeaway
The Lazard survey is a mirror. It shows us that the software industry’s valuation paradigm is shifting from growth to moat quality. Crypto is next. The current sideways market is not a pause—it is a filtering process. Capital is voting with its feet. The protocols that survive will be those that can answer the same question the Lazard investors are asking: what is your proprietary data, and how does your network compound?
For the rest, the silence will be absolute.