Author: Meng Yan's Thoughts on Blockchain

What advantages does blockchain have over mobile payments? This has been a question that many blockchain experts found difficult to answer. In 2015, the renowned blockchain evangelist Andreas Antonopoulos was challenged during a speech by an audience member who asked, "Mobile payments are convenient and fast, while Bitcoin is slow and troublesome. What advantages does it have in payments?" He did not take the defensive route; instead, he painted a vivid scenario: if a self-driving taxi operates independently, needing to collect payments and pay for charging, could Bitcoin be a more ideal payment solution?

This is a thought-provoking question. In fact, for over a decade, many thinkers in the blockchain field, such as Zhu Jiaming and Xiao Feng, have contemplated this issue and posed a bold question: perhaps blockchain was not designed with humans in mind from the very beginning, but rather for AI and robots.

Ten years ago, discussions about AI agents and autonomous driving were still trendy research in laboratories, but now these technologies are rapidly being deployed. Currently, the development of multi-agent AI organizations has become the hottest direction—will it create a new stronghold for blockchain applications and ultimately facilitate the combination of AI and blockchain?

Multi-Agent Systems Are Becoming the Hottest AI Application Direction

Starting in the second half of 2025, multi-agent systems suddenly became one of the hottest directions for AI to land. A significant proportion of the most attention-grabbing projects in the AI field belong to this direction.

Leading the Agentic AI field, Anthropic has released products like Cowork, Agent Teams, and Managed Agents, clearly indicating their leadership intent in this direction.

Google’s Agent Development Kit (ADK) offers a standardized framework that helps developers quickly build hierarchical and scalable multi-agent systems, supporting parallel, sequential, and cyclic orchestration.

OpenClaw allows ordinary users to deploy a "Lobster Team" on their local machines to collaboratively complete multi-step workflows, sparking intense discussions about "AI employees" and "one-person companies" among the public.

ByteDance's DeerFlow 2.0 quickly topped GitHub Trending after being open-sourced; it is a super-agent runtime infrastructure capable of orchestrating sub-agents, long-term memory, and Docker sandboxes, autonomously completing long and complex tasks ranging from a few minutes to several hours, thereby completely addressing the traditional AI "manual handover" pain point.

YC's CEO Gary Tan released gstack, transforming Claude Code into a virtual startup team equipped with roles such as CEO, designer, engineer, QA, security officer, among 23 professional roles, allowing one person to simulate the operation of an entire startup;

The TradingAgents project launched jointly by UCLA and MIT teams allows multiple agents to take on roles such as fundamental analysts, sentiment analysts, technical analysts, and risk managers, making investment decisions through debates and collaboration, simulating the organizational processes of real trading companies;

Paperclips.AI focuses on organizing a group of agents into a complete structure that can autonomously operate a company, including organizational charts, budgets, governance, and goal setting, achieving a zero-human-intervention business closed loop.

These projects point towards a clear trend: people are no longer satisfied with single AI assistants, but are beginning to build real work teams and collaborative networks using multi-agent architectures. In the past, we primarily used AI to "ask questions and receive answers," but now we are starting to have AI "team up to work"—collaborating, supervising each other, making autonomous decisions, and completing complex tasks that human organizations can handle.

This is not just a simple tool upgrade; it is a key turning point for AI transitioning from a personal efficiency tool to an organizational-level technology. The core of multi-agent systems (multi-agent system) is that multiple agents are no longer isolated "role players," but form a dynamic network that can create accounts, scale flexibly, recombine, and even collaborate across organizational boundaries based on task demands. They can handle high-frequency, minute, cross-entity, and cross-jurisdiction value exchanges while triggering complex contract execution conditions. These characteristics require multi-agent organizations to have an entirely new payment and trading infrastructure.

Why Must Multi-Agent Applications Be Equipped with a New Payment System?

When AI multi-agent organizations move from laboratories to real tasks, payments and value exchanges are no longer optional features but the lifeblood of system operations.

Once a multi-agent network begins processing actual business, it will generate a substantial number of high-frequency, small, cross-entity, and even cross-jurisdiction payment demands. Agent A may complete a content generation in seconds, after which Agent B immediately needs to pay for its usage model; Agent C, after processing logistics data, must immediately pay Agent D for data usage fees; during multinational cooperation, an agent from Singapore may need to pay an agent on a US server for computational resources. These payment frequencies could reach dozens of times per minute, with amounts as small as 0.1 cents, and the participating parties may be entirely different organizations or individuals. These exchanges are often accompanied by complex value flows: not only money, but also data, computational power, model invocation rights, fragments of intellectual property, etc.

More importantly, the dynamism of multi-agent organizations far exceeds that of traditional organizations. Agent accounts can be created at any time, organizations can expand and recombine at any time. The task in the previous minute may require a small team of five agents, while the next task a minute later may need twenty agents to instantly reconfigure, some of which may come from external partners. When agents across organizations pay each other, it inevitably involves triggering complex conditions for contract execution: payments only automatically execute when Agent A completes a certain verification, Agent B delivers specific data, and an external oracle confirms a market price reaches a threshold. These conditions may be nested, multi-layered, and real-time, which traditional contracts cannot describe at all, and the traditional banking system cannot respond in real time.

Traditional banking systems are incapable of fulfilling this scenario. They are accustomed to a net settlement model that relies on large amounts, batch processing, and manual reviews, with response times calculated in hours or even days. They cannot provide instant account opening services for tens of thousands of agents, cannot handle the requirement that each transaction must have complex condition triggers, and cannot provide that kind of "personalized service"—agents need code-level, millisecond-level automatic execution, instead of calling customer service or submitting paper applications. The rules of banks are designed for people, and the processes are designed for stable institutions. When faced with an agent network that can instantly reconfigure, never sleeps, and is globally distributed, its ceiling is immediately exposed.

The advantages of blockchain as a new generation of financial market infrastructure are highlighted in this scenario.

Blockchain is essentially a distributed ledger that allows all participants to share a real-time updated public ledger without the need for repeated reconciliations. Smart contracts write contract terms directly into code, automatically executing once the conditions are met without third-party intervention. Programmability makes payments no longer just simple transfers; they can be automated processes embedded with any complex logic: triggering upon condition fulfillment, atomic execution, and rollback upon failure. One-to-one gross settlement replaces traditional netting, with each transaction completing clearing and settlement simultaneously as confirmation occurs. Atomic settlement ensures that value transfer and asset delivery happen simultaneously, preventing any party from defaulting. Instant finality means that once a transaction is on the chain, it is irreversible and tamper-proof.

These features align almost perfectly with the operational logic of multi-agent organizations. Agents need to create accounts anytime, while the cost of generating blockchain addresses is nearly zero; agent organizations must scale flexibly, and smart contracts can deploy new rules instantly; cross-organization collaboration requires complex condition triggers, for which smart contracts are inherently built; high-frequency micro-payments need low-cost instant arrival, while blockchain's Gas fees and Layer 2 solutions are driving costs down to negligible levels. Traditional infrastructure is centralized, rigid, and slow, while blockchain is decentralized, flexible, and real-time.

We are increasingly seeing that multi-agent organizations are not just simply piecing together AI tools, but constructing an entirely new collaborative paradigm. This paradigm poses unprecedented requirements for payment and trading infrastructure, and blockchain is currently the only mature technological system capable of meeting these demands. It is not merely an enhancement but a necessary infrastructure. When AI agents begin to work together in earnest, blockchain will no longer be optional, but essential.

Multi-Agent Applications Will Become the “Base” for Blockchain Payments

In traditional C2C payment scenarios, blockchain has not shown outstanding performance. When ordinary people transfer funds, WeChat or Alipay can complete the process in seconds with just a few taps to input the amount and confirm via scanning a code. However, a blockchain wallet requires copying addresses, verifying Gas fees, and waiting for block confirmations, which clearly lags in user experience. Over the past decade, blockchain has struggled to compete with mobile payments in day-to-day small transfers and face-to-face payments led by humans.

However, in high-frequency, automated, contract-driven payment scenarios between AI agents, the advantages of blockchain are significantly ahead.

Agents do not need QR codes, nor do they require human confirmation. They require payments to be automatically executed. As soon as preset conditions are met, the smart contract triggers the transfer without any intermediary intervention. Programmability allows payments to embed complex logic: funds will only be released when Agent A delivers specified content, Agent B completes data verification, and an external oracle confirms that the market price reaches a threshold. If any step fails, the transaction will automatically roll back. Blockchain supports 24/7 uninterrupted operation, with instant arrival between any global addresses, and every transaction carries instant finality. These capabilities are currently unattainable by traditional banking systems and mobile payment platforms.

The applications of multi-agent organizations will become the “main field” for blockchain payments.

Initially, mobile payments had no apparent advantages in face-to-face transactions. At that time, people were still accustomed to cash and card payments, and mobile payments appeared redundant even in small shops. However, it found a breakthrough in e-commerce scenarios. Order payments on platforms like Taobao and JD required online instant settlement with support for massive concurrency, allowing mobile payments to quickly establish a foothold. It first honed the e-commerce payment experience to perfection, accumulating users, merchants, and network effects, subsequently benefiting society as a whole. Today, when we scan a code to pay, it's because mobile payments first succeeded in that e-commerce stronghold.

AI multi-agent organizations will become the most solid and explosive stronghold for blockchain payments and value exchanges.

Here, payments are no longer occasional human actions, but rather the norm in system operations. They can be micro-payments, computational resource rental fees, model invocation fees, data usage fees, or intellectual property shares occurring every second. These payments require complex conditions embedded and atomic execution. Traditional payment infrastructures struggle to cope, whereas blockchain naturally adapts. It does not need to change user habits, as agents themselves are code. It does not need customer service support, as everything is guaranteed by contracts. It does not need centralized risk control, as trust is provided by cryptography and distributed consensus.

I believe this precisely reflects the penetrating nature of blockchain technology. It establishes an irreplaceable structural advantage in the scenarios that need it most.

Blockchain does not need to fully replace existing payment systems. It only needs to first take root in areas where humans temporarily cannot make use of it, establishing a brand new value network. When AI agents begin to work in large groups, this network will rapidly grow, extending from micro-payments between agents to broader economic activities, ultimately benefiting human society. Mobile payments proved themselves through e-commerce, while blockchain will prove itself through AI multi-agent organizations.

When the stronghold of agent payments is established, the position of blockchain in the entire digital economy will also change drastically.

New Form of AI Economy

The combination of AI multi-agent organizations and blockchain is far more than just a technical overlay. It will open a completely new economic landscape, profoundly changing resource allocation, social exchanges, individual income, and the innovation ecosystem. Let us analyze this from four dimensions.

First, significantly enhance the overall performance and resource allocation efficiency of AI multi-agent systems.

Currently, the vast majority of multi-agent applications are still in a "playing house" phase. Developers mainly rely on agent skills, hooks, MCP, and prompt engineering methods to simulate and customize personalized "digital employees." This is essentially still a primitive state of role-playing where everyone is crafting seemingly professional AI roles with prompts and then letting them chat and divide tasks to simulate a multi-step workflow. While it appears lively, the actual effectiveness compared to using a single, all-purpose AI assistant is quite limited.

True multi-agent organizations are completely different. Some agents will possess unique resources and capabilities that cannot be easily mimicked or substituted through simple customization. These capabilities may include proprietary datasets, exclusive model weights, real-time data sources in specific fields, high-precision simulation environments, or industry experience accumulated over long training periods. They can only be developed, nurtured, and deployed by organizations with unique resources absorbing costs. Calling upon these advanced agents inevitably involves real payments.

Smart contracts on blockchain play a key role here. They can define calling rules, pricing mechanisms, quality verification, and fee settlement all in code. Once the conditions are met, payments automatically trigger, and resources are automatically delivered; when conditions are not met, funds automatically roll back. The entire process is efficient, secure, programmable, and auditable. The past inefficiencies of relying on manual negotiation, email confirmations, and post-fact reconciliation will completely vanish. Resource allocation efficiency will thus significantly improve, and the overall performance of the multi-agent system will reach a new level. This is not a simple cost reduction, but a genuine expansion of the system's capability boundaries.

Second, greatly promote the scale of social exchanges and the speed of economic growth.

Traditional financial infrastructures set very high barriers for micro, frequent transactions with complex conditions. Banks impose minimum transfer limits, clearing has time windows, and international payments incur exchange rate and compliance costs. These frictions exclude a large number of potential transactions.

When blockchain provides a low-friction micro-payment and value exchange network for AI agents, the situation will change fundamentally. Agents can easily complete computational resource rentals for a few cents, data calls, model fine-tuning services, or even instant settlement of individual API calls. Transactions that were previously suppressed due to high costs will now become feasible. A massive volume of previously impossible exchanges will be unleashed, and the speed and scale of economic cycles will significantly accelerate.

Imagine: a content creation agent pays a small copyright fee to the source-providing agent each time it generates a piece of high-quality text; an investment analysis agent pays the data source agent each time it invokes real-time market data; a logistics optimization agent pays the map service agent for each completed path planning task. These micro-payments, when accumulated, will form an extremely vast network of value flow. The density and frequency of economic activity will increase, and overall economic growth will gain new momentum.

Third, enable ordinary people to truly earn income through AI agents, addressing the structural gap in supply and demand in the AI era.

One of the most prominent contradictions in the AI era is that large model companies possess core capabilities, while the needs and supply of many ordinary people struggle to connect effectively. Many possess unique data, experiences, or scenarios but lack the capability to translate them into AI services; meanwhile, there are numerous tasks requiring specialized agents that cannot find suitable service providers.

The one-person multi-agent model will become a new form of employment. Ordinary individuals can deploy and operate their own agent networks, encapsulating their knowledge, data, or industry insights into callable agent modules, and then offering services externally through blockchain networks while automatically receiving payments. Some may be skilled in local life services and can train localized life assistant agents; others familiar with niche fields may develop specialized analytical agents for those vertical domains. These agents are no longer free toys but can independently create income as economic units.

Consequently, a new balancing mechanism for supply and demand in the AI era will emerge. The supply side will no longer be dominated solely by a few large companies, and the demand side will be able to precisely reach the most suitable agent services through micro-payments. Ordinary people will no longer just be consumers of AI but can become contributors and beneficiaries of the AI value network. This will significantly alleviate the employment pressures brought about by AI, while also better unleashing the innovative vitality of the entire society.

Fourth, prevent large AI model companies from evolving into new economic oligopolies.

Currently, there exists a severe imbalance in the AI ecosystem. All the knowledge and experience encapsulated in prompt engineering, skills, and other techniques exist almost entirely in free and open-source forms, making it difficult to gain sustained economic incentives. Developers are racing to burn tokens, receiving only cheap applause on social networks, and rarely translating that into income. Only large model companies have clear business models since they control the underlying computational power and foundational models.

More dangerously, once large model companies observe a successful model, they often only need to follow up slightly at the model level to easily replicate or even surpass the original AI startup. Numerous innovation teams are thus rapidly eliminated, and the innovation ecosystem faces the risk of being harvested. When Anthropic launched its Managed Agents product in early April, some lamented that at least 1,000 startups woke up one morning to find their value reduced to zero.

When agents themselves become economic elements capable of autonomously receiving payments, the situation will change. The value network will undergo a decentralized reconstruction. Each agent can independently set prices, settle accounts, and accumulate reputation and assets through blockchain. Successful agents will no longer depend on a specific model, but rather become independent nodes in the network. Developers can continually iterate on their agents to earn direct income without having to relinquish all value to the underlying model providers.

Large model companies will still remain important, but they will transition from being rule-makers to infrastructure providers. Their overwhelming advantages will be effectively balanced, and the diversity and vitality of the innovation ecosystem will be safeguarded. This is not a negation of large model companies, but it will make the entire AI economic system healthier and more sustainable.

The integration of AI and blockchain is sketching an unprecedented economic landscape for us. In this landscape, resource allocation will be more precise, exchanges will be smoother, ordinary people will have new sources of income, and the innovation ecosystem will maintain openness and vitality. The dividends brought by this integration go far beyond the technology itself. It will profoundly impact our modes of production and patterns of wealth distribution in the next twenty years.

Short-term Barriers and Practical Prudence

Although the trend is clear, the process of implementation will not be smooth. The integration of AI multi-agent organizations and blockchain faces several real and tricky barriers. We must confront these to avoid blind optimism.

First, while the United States has made strides in digital asset legislation, it has not been fully realized. The CLARITY Act is struggling to make headway, and the Treasury and regulatory agencies are still pushing forward with details regarding anti-money laundering, reserve asset management, and other issues, facing continued resistance from traditional powers regarding stablecoin issuance and smart contract payments. Perfecting the regulatory framework takes time, and short-term uncertainties in execution will still restrict large-scale applications.

Secondly, other countries, including China, still exhibit a conflicted regulatory attitude. Many economies are concerned about monetary sovereignty, capital flows, and financial stability issues, making it difficult in the short term to quickly formulate clear and friendly frameworks. Legislative delays and policy fluctuations create additional friction for cross-jurisdiction cooperation among agents.

Thirdly, AI practitioners generally lack an understanding of blockchain and harbor cognitive biases. In the AI community, blockchain is often simplistically equated with speculation or even fraud. Many developers only see on-chain Gas fees and confirmation delays but rarely delve into understanding the structural value of distributed ledgers and smart contracts for multi-agent organizations. This cognitive gap leads to excellent AI talents hesitating to invest in related fields, resulting in slow progress for integration projects.

Simultaneously, the blockchain industry is still in a low point regarding funding, talent, and confidence. Shadows from past years' speculative bubbles bursting, alongside fraud incidents and project failures, remain. High-quality projects find it difficult to secure funding, and there is serious talent loss; the industry's overall confidence will take time to rebuild. If not handled properly, these issues can amplify external biases, further slowing down the pace of integration with AI.

These barriers are real, and are unlikely to be completely eliminated in the short term. They remind us that any significant technological integration does not progress linearly, but instead requires repeated breakthroughs at the cognitive, regulatory, and practical levels.

However, precisely because these barriers exist, the strategic significance of this integration is highlighted. Whoever can first overcome cognitive bottlenecks, proactively fill regulatory gaps, and invest resources to nurture cross-disciplinary talent will be poised to seize the initiative in the new paradigm of AI and blockchain integration.

Returning to the essence of technology, casting aside short-term noise, and investing in genuine cognitive upgrades and practical explorations is the way to go. History repeatedly demonstrates that at major turning points in technology, hesitant onlookers often miss the window, while those who dare to confront obstacles and continue to iterate ultimately become the forces propelling the wave.

The trend is set.

The development of AI multi-agent organizations will inevitably lead to a deep integration of AI and blockchain. This is not a possibility, but a historical inevitability driven by technological logic. Agents need high-frequency, micro, and complex condition-triggered automatic payments; blockchain precisely provides the most suitable infrastructure.

The United States has a relative edge due to its digital asset legislation, already accumulating significant first-mover advantages in stablecoins and on-chain infrastructures. In the coming years, they are likely to transform this integration into real productivity and economic control. China must not be complacent in this integration. Given the scale of talent, technological reserves, and resource investments in AI core technologies between China and the United States, there is no large gap in pure competition over models and applications. If, several years from now, the AI ecosystem in China shows a significant disparity compared to that of the USA, it will surely be because the US has undertaken initiatives that China could not or found challenging to pursue. Currently, blockchain payments may be something along those lines.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。