Gabriela Santos, a global market strategist at JPMorgan, recently advised investors to diversify their AI holdings across regions and industries. On the surface, this is standard portfolio theory—don't put all eggs in one basket. But for those of us who have spent the last decade inside the blockchain ecosystem, her words carry a deeper resonance. They echo the very principle that underpins our movement: decentralized distribution of trust and value. When a pillar of centralized finance starts preaching diversification, it is not just a risk management tactic. It is an implicit admission that the most valuable networks are those that resist concentration.

Let me begin with a confession. In 2017, I spent three months auditing the whitepapers of 42 failed ICOs. Every single one of them had a centralized flaw—a single point of failure in governance, a founder with too many tokens, a smart contract that could be paused by a single key. I published those findings in a manifesto I called "The Soul of the Chain," arguing that decentralization is not a technical feature but an ethical imperative. The market laughed at me then. Today, JPMorgan is essentially arguing the same thing about AI investments. The parallel is not accidental.

Context: The JPMorgan Signal

Santos's recommendation is not a throwaway line. She is a global market strategist for one of the world's largest asset managers. When she says diversify AI holdings, she is speaking to pension funds, endowments, and sovereign wealth funds. Her rationale is straightforward: the AI industry is no longer a single bet on NVIDIA or OpenAI. It has splintered into chips, models, applications, data, and services, each with its own growth cycle and risk profile. Moreover, regional differences—regulatory frameworks in Europe, manufacturing scale in China, research depth in the US—create distinct opportunities that no single stock can capture.

From a Web3 perspective, this is a beautiful analogy. Just as Ethereum once dominated the smart contract narrative but now coexists with Solana, Avalanche, and dozens of L2s, AI capital is beginning to flow into a multi-chain world. The mistake would be to assume that one foundation model or one chipmaker will win everything. The blockchain industry learned that lesson the hard way in 2018 when every project claimed to be the "Ethereum killer." The survivors were those that embraced interoperability, not conquest.

Core: The Seven Dimensions of AI Decentralization

I have spent the past week reverse-engineering Santos's argument through the lens of my own analytical framework—a seven-dimensional model I developed while advising institutional allocators on blockchain exposure. Let me walk you through each dimension and show how the same logic applies to both AI and crypto.

1. Technical Route Diversification

Santos implicitly assumes that no single AI technology will dominate. This is the same assumption that drives the multichain thesis. In blockchain, we have proof-of-work, proof-of-stake, DAGs, and sharding. In AI, we have transformers, state-space models, and hybrid architectures. A diversified portfolio hedges against the risk that a paradigm shift makes today's leading model obsolete. I saw this firsthand in 2022 when I studied zero-knowledge proofs for my MS thesis. The technology that was dismissed as too slow in 2020 became the backbone of privacy in 2024. The same will happen with AI models. Don't confuse liquidity with loyalty.

2. Commercialization Stage

The AI market is moving from infrastructure build-out to application penetration. This mirrors the shift from DeFi summer 2020 to the NFT and gaming explosion of 2021. The first movers in infrastructure capture outsized returns, but the real value accrues to the applications that reach mainstream users. Santos's diversification call is a signal that the infrastructure phase is mature. The next phase belongs to vertical AI solutions—healthcare, finance, manufacturing—just as the next phase of Web3 belongs to real-world asset tokenization and decentralized physical infrastructure networks (DePIN).

3. Industry Impact

Diversification across industries is not just about risk; it is about capturing the spread of AI adoption. The same logic applies to blockchain: DeFi, supply chain, identity, and gaming all have different adoption curves. I have written extensively about the "ethical node" concept—the idea that resilience comes from a network of diverse participants. A single industry collapse should not bring down the entire portfolio. This is the same reasoning that leads Web3 founders to build on multiple chains.

4. Competitive Landscape

Santos's advice implicitly acknowledges that the AI competition is no longer a winner-take-all game. In blockchain, we have seen the same evolution. In 2020, everyone thought Ethereum would capture all value. By 2024, the market cap is distributed across dozens of L1s and L2s. The reason is simple: each chain optimizes for a different trade-off—security, speed, cost, privacy. AI models will do the same. Some will specialize in code generation, others in medical imaging, others in customer service. A diversified portfolio is the only rational response to a fragmented competitive landscape.

5. Ethics and Safety

This is the dimension where Web3's philosophy aligns most powerfully with Santos's recommendation. Centralized control of AI—whether by a single company or a single country—creates systemic risk. If that central node is compromised, the entire system fails. The same is true of blockchain. The DAO hack in 2016, the FTX collapse in 2022—each was a failure of concentration. Diversification is an ethical hedge. It distributes power, reduces the risk of bias amplification, and makes governance more resilient. I have argued for years that decentralization is an ethical imperative. Santos's diversification advice is a financial translation of that same principle.

6. Investment and Valuation

Here is the hard truth: AI valuations have expanded to the point where individual stock picking is dangerous. The same is true for crypto. In 2021, everyone thought they could pick the next 100x token. Most got burned. The smart money moved to index funds and diversified portfolios. Santos is saying the same about AI. The era of easy alpha is over. The next phase requires diversification across regions, industries, and asset classes. This is not a bearish call; it is a maturity call. The market is telling us that the low-hanging fruit has been picked. Now we need to farm the entire orchard.

7. Infrastructure and Compute

Finally, the diversification thesis rests on the assumption that compute is becoming abundant and distributed. The same is true for blockchain. We are moving from a world where only a few miners could secure the network to a world of staking pools, rollups, and shared security. In AI, the cost of inference is dropping exponentially, enabling edge computing and decentralized inference networks. This is where Web3 and AI converge. Projects like Bittensor, Render Network, and Akash are already building decentralized compute marketplaces. Santos's advice is a green light for institutional capital to explore these infrastructure plays.

Contrarian: The Risks of Pseudodiversification

Now let me play the contrarian, because that is what the framework demands. Diversification is not a silver bullet. In fact, it can be a trap. I have seen many investors buy a basket of AI stocks only to realize they all depend on NVIDIA's GPU supply. That is not diversification; it is correlated risk. The same happens in crypto: a portfolio of five different DeFi tokens may all collapse if Ethereum gas fees spike. Diversification only works if the underlying assets are truly uncorrelated.

Santos does not address this. She speaks of regions and industries but does not provide a framework for measuring correlation. As a blockchain auditor, I have learned that the key is to look at the value chain. If the same risk factor—say, government regulation of AI—affects all your holdings, you are not diversified. You are just holding a larger basket of the same fragility.

Another blind spot: diversification can dilute conviction. The best investors in history—Buffett, Thiel, Sequoia—made concentrated bets. They did not spread their capital across fifty companies. They found the one that would define the decade and went all in. Santos's advice is for the average institutional investor, not the visionary. But for the Web3 community, which is built on conviction, the call for diversification might feel like a betrayal of the thesis. If you truly believe in decentralization, should you not put all your capital into the most decentralized protocol?

My answer is no. Even the most decentralized ecosystem needs multiple nodes. The blockchain itself is a diversification of trust across validators. Similarly, an AI portfolio should be diversified across layers of the stack. The key is to ensure that each holding is independently viable. I have a rule: never invest in a project that cannot survive without the success of another project in your portfolio. That is the true test of diversification.

Takeaway: The Vision Forward

So what does this mean for the Web3 community? It means that the next frontier of institutional capital is not just AI or just crypto—it is the intersection. JPMorgan is telling us that AI is becoming a multi-asset ecosystem. And the only way to manage that ecosystem at scale is through decentralized infrastructure. Smart contracts, oracles, DAOs, and tokenized assets will be the plumbing that enables cross-regional, cross-industry AI investment. We are not just building an alternative financial system; we are building the rails for the AI economy.

I will leave you with a question: If the world's largest bank is now preaching diversification, what does that say about the future of centralized power? The answer is clear. The era of monolithic trust is ending. Whether it is AI or blockchain, the future belongs to networks that distribute value, power, and risk across many nodes. Don't confuse liquidity with loyalty. The real wealth is in the network itself.

This article is not investment advice. It is a reflection on why a strategy report from a Wall Street strategist feels like a validation of everything we have been building in the Web3 space. The seeds we planted in 2017 are now being harvested by the very institutions we once sought to disrupt. And that, I believe, is the most beautiful irony of all.