Written by: Tiger Research

Translated by: AididiaoJP, Foresight News

Fear of missing out is sweeping through the cryptocurrency industry. From exchanges to security companies, various institutions are launching AI-driven services. This article will explore the reasons companies are choosing to make this arrangement at the current time.

Key Points

• Cryptocurrency companies covering exchanges, security, payments, research, and other fields are simultaneously launching AI-related services.
• Unlike previous cycles, the current leaders are mature and profitable companies like Coinbase and Binance. Artificial intelligence has evolved from conceptual hype into a necessary operational capability.
• The motivations for different industries adopting AI vary: exchanges aim to reduce user churn; security companies strive to address audit blind spots; payment infrastructures focus on the emerging agent economy.
• There is a gap between feature launch and actual application. The fear of missing out on AI and competitive pressure are driving companies to accelerate their layout at a pace exceeding actual demand.
• Real demand and competitive anxiety are jointly driving this wave. The core issue is how to distinguish truly value-creating applications from superficial branding actions.

Cryptocurrency Companies Are Launching AI Services

Artificial intelligence is currently the most attention-grabbing field in the global market. General-purpose tools like ChatGPT and Claude have integrated into daily life, while platforms like OpenClaw further lower the technical barriers to building agents.

Although the cryptocurrency industry has responded slightly slower during this wave, it is currently accelerating the integration of AI capabilities across various verticals.

What specific AI services have these companies launched? What motivates them to enter this field?

How Cryptocurrency Companies Apply AI

Research Field

Source: Surf AI

The cryptocurrency research field has structural issues: on-chain data, market sentiment, and key indicators are scattered across different platforms, making validation difficult. General-purpose AI often provides inaccurate answers when addressing cryptocurrency-related issues.

In response to this reality, projects like Surf have launched dedicated AI research tools for the cryptocurrency field, integrating dispersed data sources. Among all AI application scenarios in the cryptocurrency field, research tools have the lowest barrier to entry for ordinary users, requiring no programming or trading experience to use.

Trading Field

Source: Bitget

Exchanges are at the forefront of AI application.

Each exchange has a different path. Some directly open proprietary trading data to users; others allow users to issue commands to AI agents using natural language, with agents handling everything from analysis to execution.

Exchanges have provided API services for many years. The current change lies in the addition of an interaction layer: through interfaces like MCP and AI Skills, non-developers can also invoke exchange functions via AI agents. Tools previously limited to developers can now be operated through natural language.

This change aligns with the evolution of user demographics. An increasing number of users without programming backgrounds are starting to use AI agents to build automated trading strategies. Users only need to describe their strategy ideas, and the agent can complete the algorithm setup and operation.

For exchanges, this trend presents both opportunities and challenges. As the user base driven by AI continues to grow, their stickiness to a single platform decreases, as agents can flexibly execute trades across different exchanges. The core motivation for exchanges to actively layout AI is to quickly attract users and enhance their activity on the platform.

Unlike information retrieval applications, trading involves real asset management, requiring higher judgment and accountability mechanisms. However, as barriers to use gradually lower, this field is also opening up to ordinary users.

Security and Audit Field

Source: Certik

Traditional smart contract audits rely on manual line-by-line code reviews, which are slow, costly, and variable in quality based on who performs them. Currently, AI has been introduced into the workflow: first, AI scans the code, and then auditors perform targeted deep reviews. This approach enhances efficiency and coverage without replacing human auditors.

CertiK is a representative company in this field. The company has faced scrutiny due to security incidents occurring after some audit projects. However, such incidents often occur outside the audit scope—audits only cover the code at specific points in time and do not include ongoing monitoring.

CertiK has addressed this gap with AI. It has introduced real-time monitoring mechanisms after audits are completed and displays this through public panels. Since the enhanced monitoring capability is driven by AI, there is no need for substantial human labor, benefiting both CertiK and its audit subjects.

In the security field, the application of AI is not intended to disrupt existing services but to expand the boundaries of human work: improving the accuracy of audits and filling monitoring blind spots in the post-audit phase. For blockchain security companies, AI is not a new business direction but a tool to solve existing business pain points.

Payment Infrastructure Field

Source: Coinbase

For AI agents to participate in economic activities, they must have available payment channels, such as paying for APIs, purchasing data, or acquiring services from other agents. For agents, the most compatible payment method is on-chain wallets paired with stablecoins.

Currently, there are mainly two models. One is a universal protocol that embeds payment functions within HTTP requests, enabling agents to complete on-chain settlements simultaneously when invoking paid APIs. The second is payment plugins for agents, allowing agents to execute payment operations only within the human-defined permissions and limits.

Payment infrastructure is the field most closely associated with stablecoins. However, since the payment entity is an AI agent rather than a natural person, there is currently no fully mature operational model.

Source: Circle

The issuer of the stablecoin USDC, Circle, is also receiving market attention. The company has released a proposal to connect its Gateway payment infrastructure with the x402 protocol and invites developers and researchers to participate in review and co-creation.

This field has yet to reach a mature phase, but the market has begun incorporating related expectations into pricing. One of the significant drivers of Circle's stock price increase is the narrative surrounding payments with AI agents. Compared to the aforementioned fields, the realization of payment infrastructure still requires a longer time, but it has established itself as one of the most important macro themes in the current market.

Why Are Cryptocurrency Companies Entering the AI Field at This Time?

When ChatGPT was launched in November 2022, both AI and the cryptocurrency industry were not yet mature. Although AI models showed certain capabilities, they were not yet reliably completing tasks; the cryptocurrency industry, meanwhile, was mired in a severe trust crisis following the FTX collapse.

Since then, there have been significant advancements in AI technology. Over the past year, the capabilities of major models have greatly improved, and their practicality has significantly increased. In contrast, the cryptocurrency industry has largely remained in the phase of "borrowing" the concept of AI, evidenced by meme coins themed around AI, AI agents lacking real functionality, and marketing-driven promotional language. Decentralized AI infrastructure projects continue to emerge, but their product quality shows significant discrepancies compared to similar native AI services.

Currently, this gap continues to widen. In the AI field, the maturity of infrastructures like MCP (which supports agents in directly calling external tools) and OpenClaw (which supports no-code agent building) has brought the era of agents from concept to reality. Cryptocurrency companies are just now starting to substantively follow up.

The key to this round of changes lies in the different actors. The leaders are no longer emerging projects that use the AI concept for brand packaging but established companies with stable revenue models—Coinbase, Binance, Bitget, and others. These companies have no motivation to use AI services as marketing gimmicks. The core driving factor for their actions is not current profits but the anxiety of lagging behind industry developments, that is, fear of missing out.

Source: FORTUNE

This sense of urgency is evident in Coinbase CEO Brian Armstrong's actions. He required all engineers to complete training on AI coding tools within a week and threatened termination for those who did not meet this requirement.

However, at the same time, it is necessary to maintain prudent judgment. For instance, in automated trading, AI agents can perform price inquiries and strategy recommendations, but how many users are actually willing to entrust funds to agents for live trading? Has the x402 protocol entered the actual application stage?

Overall, the cryptocurrency industry's layout concerning AI is not a chase for short-term trends. As the contours of the AI era gradually become clearer, companies are speeding up their actions to solidify their industry positions. There remains a gap between feature launch and real application, but the identity of the acting entities itself carries significant implications.

One might compare the AI industry to a pool that is accumulating water. Among the early entrants, many only pretended to swim. The current entrants are seasoned players with deep reserves. How high the water level will rise, and whether the pool will expand into an ocean, remains to be seen. But one thing is certain: the cryptocurrency industry will not be marginalized in this wave.