Visa Vice President and Head of Cryptocurrency Business Cuy Sheffield recently published an article analyzing eight key directions for the evolution of cryptocurrencies and AI by 2026. He pointed out that the most important shifts in these two major fields are no longer “theoretically feasible,” but “reliably implementable in practice.” In the next phase, “reliability,” “governance capability,” and “distribution capability” will become more critical competitive dimensions than technological novelty, while programmable currencies (stablecoins) will foster new AI intelligent agent payment flows.
(Background summary: Visa announced support for four stablecoins, which can be cross-chain exchanged for legal currencies in 25 countries)
(Additional context: The hottest recent AI Agent track explained: meme coins, issuance platforms, and infrastructure)

Table of Contents

• Cryptocurrencies: From speculative assets to high-quality technological infrastructure
• AI Agents: From “intelligence level” to “trustworthiness”
• Programmable currencies foster new intelligent agent payment flows

As cryptocurrencies and AI gradually mature, the most significant shifts in these two fields are no longer “theoretically feasible,” but “reliably implementable in practice.” Visa’s Head of Crypto Business, Cuy Sheffield, pointed out that both technologies have now crossed critical thresholds, with performance significantly improved, but their actual adoption remains uneven. The core developments expected by 2026 stem from this gap between “performance and adoption.”

Cryptocurrencies: From speculative assets to high-quality technological infrastructure

Cuy Sheffield believes that the first decade of cryptocurrency development was characterized by “speculative advantage”—its market is global, continuous, and highly open. The intense volatility made crypto trading more dynamic and attractive than traditional financial markets. However, during this period, early blockchain networks were slow, costly, and lacked stability. Aside from speculative scenarios, cryptocurrencies rarely surpassed existing traditional systems in terms of cost, speed, or convenience.

Today, this imbalance is beginning to reverse. Blockchain technology has become faster, more economical, and more reliable. The most attractive use cases for cryptocurrencies are no longer speculation but infrastructure-related—especially in settlement and payments. As cryptocurrencies become more mature, the central role of speculation will gradually weaken: it won’t disappear entirely but will no longer be the primary source of value.

Stablecoins are a clear result of cryptocurrencies’ focus on “practical utility.” Their success is based on concrete, objective standards: in specific scenarios, stablecoins are faster, cheaper, and more widely accessible than traditional payment channels, while seamlessly integrating into modern software systems. Stablecoins do not require users to see cryptocurrencies as an “ideology” to believe in; their applications often occur “implicitly” within existing products and workflows—this helps organizations and companies that previously viewed the crypto ecosystem as “too volatile and opaque” to finally understand its value.

Once cryptocurrencies become part of the infrastructure, “distribution capability” becomes more important than “technological novelty.” In the past, new tokens could naturally accumulate liquidity and attention simply by existing. Now, as cryptocurrencies are embedded into payment processes, platforms, and enterprise systems, end-users often remain unaware of their presence. This shift benefits two main groups: companies with existing distribution channels and reliable customer relationships; and institutions with regulatory approval, compliance systems, and risk management infrastructure.

AI Agents: From “intelligence level” to “trustworthiness”

In the AI field, Cuy Sheffield pointed out that the practicality of agents is increasingly evident, but their role is often misunderstood: the most successful agents are not “autonomous decision-makers,” but “tools that reduce coordination costs in workflows.” Historically, this was most apparent in software development—AI agent tools accelerated coding, debugging, refactoring, and environment setup. Recently, this “tool value” has expanded significantly into more fields.

Take Claude Code and similar tools as examples. Although positioned as “developer tools,” their rapid adoption reflects a deeper trend: AI systems are becoming “interfaces for knowledge work,” not just limited to programming. Users are beginning to apply “AI-driven workflows” to research, analysis, writing, planning, data processing, and operations—tasks that are more aligned with “general professional work” rather than traditional coding.

The key point is that the intelligence level of AI models has rapidly improved. Today, the limiting factors are no longer “language fluency or reasoning ability,” but “reliability in real-world systems.” Production environments demand zero tolerance for three issues: AI “hallucinations” (generating false information), inconsistent outputs, and opaque failure modes. When AI involves customer service, fund transfers, or compliance, “roughly correct” results are no longer acceptable. Building “trust” requires four foundational elements: traceability of results, memory capabilities, verifiability, and the ability to proactively expose “uncertainty.”

Programmable currencies foster new intelligent agent payment flows

As AI systems play roles in workflows, their demand for “economic interactions” increases—such as paying for services, calling APIs, rewarding other agents, or settling “usage-based interaction fees.” This has renewed attention on stablecoins: seen as “machine-native currencies,” they are programmable, auditable, and capable of transferring funds without human intervention.

Take x402 and similar “developer-oriented protocols” as examples. Although still in early experimental stages, their direction is clear: payment flows will operate via “APIs” rather than traditional “checkout pages”—enabling “continuous, fine-grained transactions” between software agents. Currently, this field remains immature: transaction sizes are small, user experience is rough, and security and permission systems are still being developed. But infrastructure innovation often begins with such “early explorations.”

Cuy Sheffield summarized that whether in cryptocurrencies or AI, early development stages favor “eye-catching concepts” and “technological novelty”; but in the next phase, “reliability,” “governance capability,” and “distribution capability” will become more critical competitive dimensions. Today, the technology itself is no longer the main limiting factor—embedding technology into actual systems is the key.

“By 2026, the defining feature will not be ‘a breakthrough technology,’ but ‘the steady accumulation of infrastructure’—these systems quietly operate while subtly reshaping value flows and work modes.”