Visa Crypto Lead: Eight Key Evolutions of Crypto and AI by 2026

Original Title: 2026 Outlook: Key Themes Shaping Crypto and AI

Original Author: Cuy Sheffield, Head of Crypto at Visa

Original Translation: Saoirse, Foresight News

As cryptocurrency and AI gradually mature, the most important shifts in these two major areas are no longer merely "theoretically possible" but rather "practically achievable." Currently, both technologies have crossed key thresholds in performance improvement, yet the rate of real-world adoption remains uneven. The core dynamics of development in 2026 stem from this performance-versus-adoption gap.

Below are several key themes I have been closely monitoring, including preliminary thoughts on the direction of technological development, areas of value accumulation, and why the ultimate winners may be drastically different from the industry pioneers.

Theme One: Cryptocurrency Transitioning from a Speculative Asset Class to Premium Technology

In the first decade of cryptocurrency development, a key feature was its "speculative advantage" — its market exhibited global, continuous, and highly open characteristics, with intense volatility making cryptocurrency trading more dynamic and attractive than traditional financial markets.

However, at the same time, its underlying technology was not yet prepared for mainstream applications: early blockchain was slow, costly, and lacked stability. Apart from speculative scenarios, cryptocurrency has almost never surpassed existing traditional systems in terms of cost, speed, or convenience.

Now, this imbalance is beginning to shift. Blockchain technology has become faster, more economical, and more reliable, and the most attractive application scenarios for cryptocurrency are no longer speculative but in the infrastructure field — especially in settlement and payment processes. As cryptocurrency evolves into a more mature technology, the core position of speculation will gradually weaken: it will not disappear entirely but will no longer be the main source of value.

Theme Two: Stablecoins are a Clear Outcome of Cryptocurrency in "Pure Pragmatism"

Unlike previous cryptocurrency narratives, stablecoins' success is based on specific, objective standards: in specific scenarios, stablecoins are faster, cheaper, more widely accessible than traditional payment channels, and can seamlessly integrate into modern software systems.

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

It can be said that stablecoins have helped the cryptocurrency space reanchor to "utility" rather than "speculation," setting a clear benchmark for "how cryptocurrency can successfully land."

Theme Three: When Cryptocurrency Becomes Infrastructure, "Distribution Capability" Is More Important Than "Technical Novelty"

In the past, when cryptocurrency mainly played a role as a "speculative tool," its "distribution" had endogeneity—new tokens would naturally accumulate liquidity and attention by just "existing."

However, as cryptocurrency becomes infrastructure, its use cases are shifting from the "market level" to the "product level": it is embedded in payment flows, platforms, and enterprise systems, often unnoticed by end users.

This shift greatly benefits two types of entities: those with existing distribution channels and reliable customer relationships, and those with regulatory licenses, compliance systems, and risk management infrastructure. Relying solely on "protocol novelty" is no longer sufficient to drive widespread cryptocurrency adoption.

Theme Four: AI Agents Have Practical Value, and Their Influence Is Extending Beyond the Coding Field

The practicality of AI agents is becoming increasingly prominent, 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 has been most evident in the software development field—agent tools have accelerated the efficiency of coding, debugging, code refactoring, and environment setup. However, in recent years, this "tool value" has been significantly expanding into more domains.

Take tools like Claude Code, for example. Despite being positioned as "developer tools," the rapid adoption of such tools reflects a deeper trend: agent systems are becoming an "interface for knowledge work," extending beyond the coding domain. Users are starting to apply "agent-driven workflows" to research, analysis, writing, planning, data processing, and operational tasks—tasks that lean more towards "general professional work" rather than traditional coding.

The real key is not the "ambient coding" itself but the core patterns behind it:

· Users delegate "goal intent" rather than "specific steps";

· Agents manage "context information" across files, tools, and tasks;

· Workflows shift from "linear progression" to "iterative, conversational."

In various types of knowledge work, agents excel at gathering context, performing defined tasks, reducing process handoffs, and accelerating iterative efficiency, but they still have shortcomings in "open-ended judgment," "responsibility attribution," and "error correction."

Therefore, most AI agents currently used in production scenarios still require "constrained, supervised, embedded systems" rather than running fully independently. The actual value of AI agents comes from the "restructuring of knowledge workflows" rather than "replacing labor" or "achieving full autonomy."

Theme Five: The bottleneck of AI has shifted from "intelligence level" to "trustworthiness"

The intelligence level of AI models has seen rapid advancement, and the current limiting factor is no longer "a single language fluency or reasoning ability" but rather "reliability in real-world systems."

Production environments have zero tolerance for three types of issues: AI "illusions" (generating false information), inconsistent output, and opaque failure modes. Once AI is involved in customer service, financial transactions, or compliance processes, "roughly correct" results are no longer acceptable.

Building "trust" requires four key foundations: traceable results, memory capabilities, verifiability, and the ability to actively expose "uncertainty." Until these capabilities are mature enough, the autonomy of AI must be restricted.

Theme Six: System engineering determines whether AI can be implemented in a production scenario

A successful AI product will view the "model" as a "component" rather than a "finished product"—its reliability stems from "architecture design" rather than "prompt word optimization."

This "architecture design" includes state management, control flow, evaluation and monitoring systems, as well as fault handling and recovery mechanisms. It is for this reason that the development of AI is increasingly approaching "traditional software engineering" rather than "cutting-edge theoretical research."

The long-term value will lean towards two main entities: system builders and platform owners who control workflows and distribution channels.

As intelligent agent tools expand from the coding domain to research, writing, analysis, and operational processes, the importance of "system engineering" will be further highlighted: knowledge work is often complex, relies on state information, and is context-intensive, making an intelligent agent that can "reliably manage memory, tools, and iterative processes" (rather than just generate output) more valuable.

Theme Seven: The contradiction between open models and centralized control triggers unresolved governance issues

As AI systems become more capable and integrate deeper into the economic domain, the question of "who owns and controls the most powerful AI models" is causing a fundamental contradiction.

On the one hand, AI research in cutting-edge fields remains "capital-intensive" and is increasingly concentrated due to the influence of "computing power access, regulatory policies, and geopolitics"; on the other hand, open-source models and tools continue to iterate and optimize under the drive for "widespread experimentation and easy deployment."

This "coexistence of centralization and openness" has raised a series of unresolved issues: reliance risk, auditability, transparency, long-term bargaining power, and control of critical infrastructure. The most likely outcome is a "hybrid model"—where cutting-edge models drive technological breakthroughs, and open or semi-open systems integrate these capabilities into "widely distributed software."

Theme Eight: Programmable Money Drives a New Smart Entity Payment Flow

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

This demand has brought stablecoins back into focus: they are seen as "machine-native currency," with programmability, auditability, and the ability to transfer without human intervention.

Take protocols like x402, for example—the current stage is still in the early experimentation phase, but its direction is clear: payment flows will operate in the form of "APIs," rather than traditional "checkout pages"—enabling "continuous, granular transactions" between software entities.

Currently, this area still seems immature: small transaction volumes, rough user experience, and evolving security and permission systems. However, infrastructure innovation often starts with such "early exploration."

It is worth noting that its significance is not "autonomous for the sake of autonomy" but "when software can transact programmatically, new economic behaviors become possible."

Conclusion

Whether it is cryptocurrency or artificial intelligence, the early stages of development tend to favor "eye-catching concepts" and "technological novelty"; in the next stage, "reliability," "governance capability," and "distribution ability" will become more critical competitive dimensions.

Today, technology itself is no longer the main limiting factor; "embedding technology into real systems" is key.

In my view, the hallmark of 2026 is not "some breakthrough technology" but the "steady accumulation of infrastructure"—these facilities are quietly reshaping "the way value flows" and "the way work is carried out."

Original Article Link