At Davos this year, the tone shifted decisively from if to how fast.

The debate is no longer about whether digital assets, AI, or on-chain infrastructure belong in the financial system. That question is settled. What senior executives are now grappling with is speed, sequencing, and survival.

Who executes first, how fast operating models can adapt, and who gets disintermediated along the way.

At the World Economic Forum in Davos, the mood was pragmatic and increasingly urgent. Regulation is firming up. Infrastructure is real. The remaining uncertainty sits squarely inside institutions themselves.

Several voices cut through the noise by focusing not on narratives, but on operating reality.

Brian Armstrong, CEO of Coinbase, was blunt. Regulation is no longer the gating factor. The advantage now lies with organizations that can build regulated crypto infrastructure faster than incumbents can rewire legacy systems.

Armstrong shared a striking anecdote. A CEO of a top ten global bank told him that crypto is now their number one strategic priority. Not because it is fashionable, but because disintermediation is real.

Stablecoins and tokenized assets allow value to move instantly. For users, this is progress. For institutions built on friction, float, and settlement delays, it is existential. Banks see the threat clearly. They know they must adapt or risk being bypassed entirely.

Jeremy Allaire, CEO of Circle, reinforced this from the payments side. Stablecoins are no longer a crypto edge case. They are becoming programmable money for the global financial system. With clearer regulation, stablecoins like USDC are increasingly viewed as enterprise grade cash infrastructure, relevant to treasury, payments, and settlement at scale.

Vlad Tenev, CEO of Robinhood, focused on tokenization. His view was that tokenized assets will collapse the distance between capital markets and consumers. Fractional access, faster settlement, and global distribution become defaults, even if end users never realize they are interacting with on chain rails.

This thinking is now echoed by large asset managers like BlackRock, where tokenization is increasingly framed as an upgrade to market plumbing rather than crypto adoption. Even institutions like the IMF now treat stablecoins and tokenized assets as inevitable, with the focus shifting to safe and durable integration.

One additional idea surfaced repeatedly. AI is no longer peripheral to finance.

As Changpeng Zhao (CZ), CEO of Binance, put it succinctly, the native currency of AI agents will be cryptocurrency. As AI systems become more autonomous, they need native, programmable ways to transact, settle, and coordinate. Traditional banking rails were not designed for machine to machine finance. On-chain infrastructure was.

If Davos shows where consensus is forming, demo days show how execution is actually unfolding.

Recently, I attended and judged the Vector Institute FastLane Demo Day at Highline Hub, which featured 13 emerging AI startups and drew a strong mix of VC investors, executives, and ecosystem partners.

While there were not many companies positioned explicitly as future of finance startups, many were building systems that sit directly underneath it. Risk modeling. Forecasting. Decision automation. Identity. Trust and verification. These capabilities increasingly define how modern financial systems operate, whether or not they carry a fintech label.

Vector is a global AI powerhouse for applied AI, partnering with 500+ companies to move research into real world deployment. It sits at the intersection of academic research, industry, and commercialization, an orientation that closely matches what financial institutions need as they move from pilots to production systems.

I also attended the Venture Scientist Demo Day at Mila which is globally recognized as an AI research institute with 1,000+ researchers and deep strengths in foundational AI, robustness, and safety. These capabilities are especially relevant in regulated environments, where explainability, resilience, and governance matter as much as performance.

That research focus is now being paired with a deliberate commercialization engine. The launch of Mila’s $100M Venture Scientist Fund is designed to translate world class AI research into globally competitive technology companies. It reflects a clear view that breakthrough research alone is not enough. Execution, capital, and structured company building are required to turn ideas into durable platforms.

This approach aligns closely with venture studio models. Studios provide the governance, execution discipline, and early market access that research driven teams often lack, while preserving the integrity of the underlying science. For institutions, this combination reduces execution risk by creating a clearer path from research to deployment.

Across both the Vector and Mila ecosystems, one theme was consistent. The most effective path from research to impact increasingly runs through structured venture building. This is where AI excellence, institutional requirements, and real world adoption begin to converge, and where the future of finance will increasingly be co-created.

This can still sound abstract, so it helps to think about it in practical terms.

Ethereum provides the financial rails. It handles ownership, settlement, and the rules that govern how money and assets move. Once those rules are defined, they are enforced automatically.

AI operates on top of those rails as the decision layer. It determines timing, allocation, risk, and optimization in real time. Together, they change how core financial functions operate.

In treasury, stablecoins allow cash to move instantly across borders. Smart contracts define where funds can go and under what conditions. AI decides when to move capital, how much liquidity to hold, and how to manage exposure. Treasury shifts from manual planning to continuous optimization.

In capital markets, assets can be issued and settled directly on-chain. Ownership is native and settlement is immediate. AI models monitor pricing, liquidity, and risk continuously. Markets become faster and more efficient without sacrificing control.

In compliance, identity and transaction data can be verified on-chain. AI systems monitor activity in real time, flag anomalies, and enforce rules automatically. Compliance moves from periodic reviews to ongoing oversight.

Put simply. Ethereum defines what is allowed to happen.
AI decides when and how it should happen.

Together, they form the operating layer for a more automated, resilient, and programmable financial system.

In early February, we will release our report, Future of Finance: Institutional Realities and the Role of Ethereum.

It is designed to help leaders and investors move from awareness to execution, grounded in direct conversations with senior financial executives and informed by what builders are actually creating.

We will also convene investors at a VNTR investor roundtable on February 3rd to explore how the next financial stack is evolving and where value will ultimately accrue.