His thesis is that two technologies that have so far evolved along parallel paths are destined to converge. On the one hand, artificial intelligence is becoming a system capable of performing intellectual work autonomously. On the other, blockchain provides the infrastructure for that work to be transformed into economic activity through programmable money, smart contracts and verifiable digital identity. These are not two independent revolutions, Allaire argues, but two sides of the same historic change.
This is the central idea of The Agentic Economy, an 89-page treatise that the founder of Circle recently published and in which he sets out what an economy will look like where artificial intelligence agents stop being mere assistants and become genuine economic actors.
The company will cease to be as we know it
The first consequence of this convergence will directly affect the company. According to Allaire, the organizational model that has dominated the economy for more than a century is beginning to lose its meaning.
Japanese bank SBI launches the first lending service with its stablecoin JPYSC
Until now, companies gathered marketing, engineering, sales, finance, human resources, customer service and legal specialists under one roof because coordinating that work within the organization was more efficient than hiring each service separately. Artificial intelligence changes that logic.
The author argues that virtually any knowledge-based activity can become a specialized agent capable of executing specific tasks. Instead of entire departments, companies will have libraries of agents trained to draft contracts, develop software, design marketing campaigns, analyze financial risks or serve customers.
As a result, the size of a company will no longer be measured solely by the number of employees and will depend on its ability to coordinate human and artificial intelligence. Allaire imagines organizations in which a single person can direct dozens or even hundreds of specialized agents, as well as small teams capable of competing with large multinationals thanks to the automation of intellectual work.
AI agents will compete to get work
The transformation will not be limited to the interior of companies. Once those capabilities can be modularized, they can also be sold in an open market. Just as today a company hires a law firm, a consultancy or a communications agency, in the future it will be able to turn directly to a specialized agent to perform a specific task.
An agent specializing in auditing, another specializing in contract negotiation, one dedicated to designing advertising campaigns or a software engineer will be able to offer their services from anywhere in the world. Clients will no longer be only people or companies: other artificial intelligence agents will also be able to hire them to complete more complex work.
In this scenario, Allaire considers that the true economic value will no longer be concentrated exclusively in large AI models, such as ChatGPT, Gemini or Claude. These will remain essential, but will tend to become a common technological infrastructure, just as cloud computing does today.
The competitive advantage will lie in the specialized agents built on those models, capable of solving very specific problems thanks to their sector knowledge, their own data and accumulated experience. In other words, the business will no longer consist solely of developing the best artificial intelligence model, but of creating the best digital workers.
Blockchain, the infrastructure where those AIs will work
For that global market to exist, more than AI is needed. If an agent must hire another service, charge for work performed, buy computing capacity, pay for licenses or access financing, it needs an economic infrastructure capable of operating without human intervention. That is precisely where Allaire places the role of blockchain.
The founder of Circle defines it as a genuine economic operating system on which money, digital assets, smart contracts and verifiable identities will be able to circulate. While artificial intelligence generates intellectual work, blockchain will allow that work to be automatically transformed into economic activity. In Allaire's vision, millions of agents will negotiate with each other at machine speed, sign smart contracts, exchange assets and settle transactions in a matter of seconds, without the need for banks, closed platforms or traditional administrative processes.
The economy would thus stop being organized exclusively around companies and evolve towards a huge network of intelligent agents capable of collaborating with each other using open infrastructures based on blockchain.
Stablecoins will be the money of AI
If blockchain constitutes the economic operating system of this new economy, stablecoins will be its main means of payment. For Allaire, stablecoins will not compete only with bank transfers. They are called to become the native money of artificial intelligence. Just as the internet needed a common protocol to exchange information, the agent economy will need a stable digital asset with which millions of artificial intelligences can pay each other automatically.
AI agents will need to move money constantly, make millions of small-value payments, hire services for just a few seconds or buy computing capacity in real time. For all of that to be possible, money will have to be as programmable as the software itself.
Unlike a person, an AI agent cannot stop to check the solvency of the bank issuing a currency or wait several days for an international transfer to be completed. The machine economy demands practically instant payments and an asset whose value is stable and predictable.
For this reason, Allaire advocates a model based on stablecoins fully backed by safe reserves, capable of maintaining parity with fiat currency and settling transactions almost in real time. In his view, these characteristics make them the most suitable monetary infrastructure for an economy in which transactions stop being carried out at human speed and are executed at machine speed.
The transformation Allaire proposes does not end with payments. It also reaches one of the pillars of the financial system: credit. Today, a significant part of the cost of granting a loan lies in assessing the borrower's solvency, a process that consumes time and resources and that, in many cases, makes small-value transactions unviable.
According to the founder of Circle, AI will drastically reduce that cost. Agents will be able to analyze large volumes of financial information in real time, while blockchain will provide a verifiable record of payments, income, assets and contracts. As a result, risk assessment will be much faster and more accurate.
AI agents as borrowers
Allaire goes even a step further by suggesting that AI agents themselves could become borrowers. Imagine, for example, an agent hired to develop a software application. Before starting work, it could request financing to acquire computing capacity and automatically repay the loan when the client paid for the completed project. In that scenario, credit would stop depending solely on the borrower's reputation and would become backed by the income generated by the work itself.
This idea connects with another of the treatise's most striking theses, which Allaire summarizes in one phrase: velocity replaces leverage. His approach is that, in an economy based on blockchain and stablecoins, money will be able to circulate and be reused almost instantly thanks to real-time settlement of transactions. Instead of depending exclusively on creating more credit to boost economic activity, the same capital will be able to finance many more transactions simply because it will move much faster.
Although he acknowledges that risks will continue to exist, Allaire argues that this model could broaden access to financing, reduce intermediation costs and automate much of the credit market.
Identity will be the basis of trust
An economy made up of millions of autonomous agents poses a fundamental problem: trust. Who is accountable if an agent breaches a contract? How do you know who is behind software operating from anywhere in the world? How do you prevent fraud? Allaire's answer involves combining blockchain with verifiable digital identity systems.
Each agent would have a cryptographic identity, a digital wallet and a set of credentials that would make it possible to know who its owner is and who is legally accountable for its actions. That information would serve to build a verifiable reputation over time and would make it easier for companies, regulators and other agents to know who they are interacting with.
For the author, autonomy cannot mean anonymity. The greater the economic role of artificial intelligence, the greater the need to identify those responsible for its decisions.
Despite the prominence he gives to artificial intelligence, Allaire does not paint a future completely dominated by machines. In his view, people will remain essential to define objectives, exercise critical judgment, assume legal responsibilities, make complex decisions and maintain the human relationships that no artificial intelligence can completely replace.
The nature of work will change
Instead of directly executing many tasks, professionals will supervise networks of intelligent agents. Their role will increasingly consist of designing processes, coordinating systems and deciding when human intervention is still necessary. Beyond technology, The Agentic Economy raises a profoundly economic and political question: if artificial intelligence agents generate a growing share of global wealth, who will own them?
For Allaire, that will be one of the great debates of the coming decades. The problem will no longer be solely the replacement of jobs, but the concentration of the ownership of intelligence. If the most advanced agents belong to a small number of companies, they will also concentrate an ever-greater share of the economic value they generate.
On the contrary, if the infrastructure remains open and based on blockchain, millions of people and companies will be able to create, own and monetize their own agents, favoring a much more distributed economy. In that vision, blockchain stops being simply a technology for moving digital assets and becomes the mechanism that can prevent the next artificial intelligence revolution from being concentrated in a few platforms.
A challenge for banks, companies and regulators
If Jeremy Allaire's proposal comes to materialize, banks will have to adapt to an economy in which a growing share of transactions will be carried out by software and not by people. Payment systems will have to process millions of small-value automatic operations, while the banking business will evolve towards new models of custody, financing and intermediation.
Companies will also have to redefine their organization. Size will no longer be measured solely by the number of employees and will depend increasingly on the ability to combine human talent with networks of specialized agents. In that context, small teams will be able to compete with large corporations thanks to the automation of much of their processes.
Regulators will not be left out either. They will have to establish new rules on digital identity, legal liability, financial supervision and consumer protection in an environment where AI agents will be able to hire services, move money, access financing and participate directly in economic activity.
A new debate
Although The Agentic Economy is presented as a personal essay and not as a corporate document from Circle, it is inevitable to relate many of its ideas to the strategy the company has been promoting for years to turn stablecoins and blockchain infrastructure into the basis of the digital financial system.
However, the scope of the text transcends the world of cryptocurrencies. More than an essay on artificial intelligence or blockchain, it proposes a framework for understanding how the economy could be organized in a world where intelligent agents are capable of working, exchanging value, accessing financing and making decisions autonomously.
Until now, the big question was what artificial intelligences could do. Allaire asks how an economy will work in which machines not only think, but also work, charge, pay, hire services, request credit and actively participate in markets?
Many of those ideas may take years to become reality or may never come to pass. But The Agentic Economy succeeds in forcing a rethink of the relationship between artificial intelligence, money and economic organization. If that scenario does come to materialize, blockchain will stop being simply the technology underpinning cryptocurrencies and become the economic infrastructure on which artificial intelligence will operate.