Visa Executive Outlines 8 Key Trends for Cryptocurrency and AI by 2026

Original Author: Cuy Sheffield, Visa's Vice President and Head of Cryptocurrency Business

Original Translation: Saoirse, Foresight News

As cryptocurrencies and AI gradually mature, the most significant transformation in these two fields is no longer about whether something is "theoretically feasible," but whether it can be "reliably implemented in practice." Currently, both technologies have crossed critical thresholds, achieving substantial performance improvements, yet their practical adoption remains uneven. The key developments of 2026 will stem precisely from this "gap between performance and adoption."

The following are several core themes I have been following for a long time, as well as my initial thoughts on the development directions of these technologies, the areas of value accumulation, and even why "the ultimate winner may end up being completely different from the industry pioneers."

The first decade of cryptocurrency development was characterized by "speculative advantage"—its market is global, continuous, and highly open, and the extreme volatility makes cryptocurrency trading more dynamic and attractive than traditional financial markets.

However, at the same time, its underlying technology was not yet ready for mainstream adoption: early blockchains were slow, expensive, and lacked stability. Apart from speculative scenarios, cryptocurrencies have rarely outperformed existing traditional systems in terms of cost, speed, or convenience.

Now, this imbalance is beginning to shift. Blockchain technology has become faster, more cost-effective, and more reliable, and the most attractive applications of cryptocurrencies are no longer speculation, but infrastructure—particularly in settlement and payment systems. As cryptocurrencies gradually evolve into more mature technologies, speculation will lose its central role: it will not disappear entirely, but it will no longer be the primary source of value.

Stablecoins differ from previous cryptocurrency narratives in that their success is based on specific, objective criteria: in certain scenarios, stablecoins are faster, cheaper, and have broader reach than traditional payment channels, while seamlessly integrating into modern software systems.

Stablecoins do not require users to regard cryptocurrencies as an "ideology" to believe in. Their applications often occur "implicitly" within existing products and workflows—this also allows institutions and enterprises that previously considered the crypto ecosystem "too volatile and insufficiently transparent" to finally clearly understand their value.

Stablecoins can be said to help cryptocurrencies re-anchor to "utility" rather than "speculation," and they set a clear benchmark for the successful adoption of "cryptocurrencies."

In the past, when cryptocurrencies mainly played the role of a "speculative tool," their "distribution" was endogenous — new tokens could naturally accumulate liquidity and attention simply by "existing."

Once cryptocurrencies become part of the infrastructure, their application scenarios are shifting from the "market level" to the "product level": they are integrated into payment processes, platforms, and enterprise systems, often remaining invisible to end users.

This transition greatly benefits two types of entities: first, businesses that already have established distribution channels and reliable customer relationships; and second, institutions equipped with regulatory licenses, compliance systems, and risk control infrastructure. Merely relying on the "novelty of the protocol" is no longer sufficient to drive the large-scale adoption of cryptocurrencies.

The practical value of AI agents is becoming increasingly evident, yet their role is often misunderstood: the most successful agents are not "autonomous decision-makers," but rather "tools that reduce coordination costs within workflows."

Historically, this trend is most evident in the field of software development—agent tools have significantly improved the efficiency of coding, debugging, code refactoring, and environment setup. However, in recent years, this "tool value" has been rapidly expanding into many more domains.

Take tools like Claude Code as an example. Although it is positioned as a "developer tool," its rapid adoption reflects a deeper trend: agent systems are becoming the "interface for knowledge work," extending far beyond the realm of programming. Users are beginning to apply "agent-driven workflows" to research, analysis, writing, planning, data processing, and operational tasks—areas that are more aligned with "general professional work" rather than traditional programming.

What is truly critical is not "atmosphere coding" itself, but the underlying core pattern:

• What the user entrusts is the "target intention," not the "specific steps";
• Agent cross-file, tool, and task management of "context information";
• The work mode shifts from "linear progression" to "iterative and dialogical."

In various knowledge-based tasks, agents excel at gathering context, performing specific tasks, reducing handoffs in processes, and accelerating iterative efficiency. However, they still have limitations in "open-ended judgment," "assignment of responsibility," and "error correction."

Therefore, most agents currently used in production scenarios still require "limited scope, supervision, and system integration," rather than fully independent operation. The actual value of agents stems from "restructuring knowledge work processes," rather than "replacing labor" or "achieving full autonomy."

The intelligence level of AI models has rapidly improved, and the current limiting factor is no longer "mere fluency in language or reasoning ability," but rather "reliability within real-world systems."

The production environment has zero tolerance for three types of issues: first, AI "hallucinations" (generating false information), second, inconsistent output results, and third, opaque failure modes. Once AI is involved in customer service, financial transactions, or compliance processes, results that are "approximately correct" are no longer acceptable.

The establishment of "trust" requires four fundamental elements: first, traceable outcomes; second, the ability to retain memory; third, verifiability; and fourth, the capacity to actively disclose "uncertainties." Before these capabilities are sufficiently mature, the autonomy of AI must be restricted.

A successful AI product treats the "model" as a "component" rather than a "finished product"—its reliability stems from "architectural design," not "prompt optimization."

The "architectural design" here includes state management, control flow, evaluation and monitoring systems, as well as failure handling and recovery mechanisms. For this reason, AI development today is becoming increasingly aligned with "traditional software engineering," rather than "cutting-edge theoretical research."

Long-term value will tilt toward two types of entities: first, system builders, and second, platform owners who control workflows and distribution channels.

As agent tools expand from coding into research, writing, analysis, and operational workflows, the importance of "systems engineering" will become even more prominent: knowledge work is often complex, state-dependent, and context-rich, making agents that can reliably manage memory, tools, and iterative processes (rather than just agents that can generate outputs) more valuable.

As AI system capabilities grow stronger and their integration with the economic sector deepens, the question of "who owns and controls the most powerful AI models" is sparking a central conflict.

On one hand, R&D in the cutting-edge areas of AI remains "capital-intensive," and is increasingly concentrated due to the influence of "access to computing power, regulatory policies, and geopolitical factors." On the other hand, open-source models and tools continue to evolve and improve, driven by "widespread experimentation" and "convenient deployment."

This pattern of "centralization coexisting with openness" has raised a series of unresolved questions: dependency risks, auditability, transparency, long-term bargaining power, and control over critical infrastructure. The most likely outcome is a "hybrid model"—where cutting-edge models drive breakthroughs in technical capabilities, while open or semi-open systems integrate these capabilities into "widely distributed software."

As AI systems play a role in workflows, their need for "economic interactions" is increasing—such as paying for services, calling APIs, compensating other agents, or settling "usage-based interaction fees."

This demand has brought "stablecoins" back into the spotlight: they are seen as "machine-native currencies," possessing programmability and auditability, and enabling transfers without human intervention.

Take x402, a "developer-oriented protocol," as an example. Although it is still in the early experimental stage, its direction is clear: payment flows will operate in the form of "APIs" rather than traditional "checkout pages"—enabling software agents to conduct "continuous and fine-grained transactions."

Currently, this field is still in its early stages: the transaction scale is small, the user experience is rough, and the security and permission systems are still being refined. However, innovations in infrastructure often begin with such "early explorations."

It is worth noting that the significance is not "autonomy for autonomy's sake," but rather "when software can execute transactions through programming, new economic activities become possible."

Whether it's cryptocurrency or artificial intelligence, the early development stages favor "attention-grabbing concepts" and "technological novelty." However, in the next phase, "reliability," "governance capabilities," and "distribution capabilities" will become more critical competitive dimensions.

Nowadays, technology itself is no longer the main limiting factor; the key lies in "integrating technology into real-world systems."

In my view, the defining characteristic of 2026 will not be "a breakthrough technology," but rather "the steady accumulation of infrastructure"—facilities that operate silently while quietly reshaping the "ways value is transferred" and "how work is conducted."