Infrastructure and platform: a strategy for the agentic economy

The "agentic economy", a sociotechnological revolution that promises to transform human to human relations into human + agent to human + agent, while not a present concrete reality, remains a vision of the future. While general-purpose artificial intelligence systems have made definitive changes in the informational and programming spaces, the ocean for the development and production of agentic consumer-grade applications is still very blue.

In the market, we see that many companies are still occupying the space of middle men between software systems and AI models, providing shells and interfaces over chatbots. In other words, whatever the generalised "agentic economy" will be, it is not fully here yet. A visit to the AGNTCon + MCPCon North America 2026 offered some insights into what is coming, what is missing from the agentic economy, and what we can to to move in that direction:

1. The imagination of Rose from The Jetsons robots helping us with daily activities is fun, but it needs to be taken two steps further by the creation of actually useful machines if this market is to grow. Though Jean Baptiste Say's Law that "supply creates demand" is old science in terms of macroeconomics, it is still a good principle for entrepreneurship. In the XIXth century, the world fairs like the Great Exhibition (London, 1851) and the Exposition Universelle (Paris 1889) showcased the marvelous uses that modern machines were capable to effect. If markets cannot recreate this sense of spectacle, it will be hard to bring about this so-called fourth (or is it the fifth already?) Industrial Revolution.
2. The simple usefulness or coolness of products is not enough to drive the real engine of growth in macroeconomics, aggregate demand. Critical examination of the growth of Big Tech and the digital economy for sales and services reveals social structures (or lack thereof) allowing companies to take shares of the market. Case in point, the extraordinary cost of high education begot the demand for low cost educational platforms like Udemy. In this sense, a question that remains open is what social and cultural structures need changing in order to bring about the development of this agentic economy.
3. Parallel to the aforementioned social infrastructure, there is a technical infrastructure to be developed and agreed upon. In Tim O'Reilly's words, the (agentic) economy is "crying out for mechanism design", that is, the set of rules that make sure that scientists, companies and the public are all playing the same game. Just as the creation of the World Wide Web had to deal with countless challenges (viz. ensuring packets go back and forth or not with TCP or UDP, providing request idempotency, cryptography), the communication between these two very distinct types of actors, humans and AI agents, will demand a set of protocols that go way beyond the current limits of chatbot interface and MCP.

On researching the challenges for the agentic economy, the Microsoft paper on the topic states, "to unlock the full economic potential of generative AI’s communication capabilities, two developments are necessary. First, consumers and businesses must widely adopt assistant and service agents...Second, these agents must be designed to programmatically interact with each other to facilitate transactions... it remains to be seen how these advances will be adopted and implemented, or constrained, given their complex interplay with and dependence on market forces". It follows that it is up to programmers to build the protocols that govern the interactions between these distinct actors, humans and AI agents, and the digital platforms that allow building the new agentic applications.

Personally, in my work as a software engineer at Telnyx, the way to be at the vanguard of this transformation is to cooperate in the development of both the physical infrastructure and the digital platform for programmatic and now agentic systems of communication. These tactics attack at least two of those missing points, and contribute to a broader strategy towards an agentic economy. Particularly, it allows companies and developers to build, deploy, and scale AI agents to perform tasks limited only by their creativity. Collectively, it creates a space where a multitude of agents of different models interact, therefore creating the opportunity for the rules of this new reality to be organically written.

Further reading: David M. Rothschild et al., "The Agentic Economy", Microsoft Research, USA. Tim O'Reilly, "The Missing Mechanism of the Agentic Economy", https://www.oreilly.com/radar/the-missing-mechanisms-of-the-agentic-economy/ C. K. Hunt, "History of Economic Thought", ch. 6 "Subjective rationalism: the economy of Bentham, Say, and Senior", University of Utah, USA. Chan Kim and Renée Mauborgne, "What is Blue Ocean Strategy" https://www.blueoceanstrategy.com/what-is-blue-ocean-strategy/