Blockchain Capital Partner: AI is rewriting the fundamental unit of labor

Author: Kinjal Shah

Compiled by: Jiahua, ChainCatcher

In 2024, Sam Altman made a bold prediction: with the rise of artificial intelligence, a billion-dollar company founded by a single individual will soon emerge.

The core shift lies in the fact that for the first time, humans can scale in the dimension that has always limited them, which is time. When intelligence is no longer constrained by the human need for sleep but is driven by tireless machines, what will our familiar "creation and construction" look like?

Imagine this scenario: one intelligent agent delegates a task to another intelligent agent, pays with USDC upon receiving the results, and the entire transaction is settled on-chain within 400 milliseconds, with no intermediaries verifying the process.

Or, an athlete authorizes their signature touchdown celebration to be recreated for a video game marketing campaign by a world model. Alternatively, a scientist pays directly to the original researchers for a rare dataset needed for an experiment.

We are much closer to this vision than most people think.

The prevailing fear in current discussions (that AI is taking away jobs) actually misses a more interesting structural question: what happens when the basic unit of labor itself changes?

Every Transition

Regarding why companies exist, Ronald Coase provided the clearest answer in his 1937 paper "The Nature of the Firm": companies will "internalize" labor when the costs of coordinating through the market exceed the costs of direct employment.

Every major labor transformation in history has been a direct result of decreasing coordination costs. As the friction of finding, paying, and managing work diminishes, the boundaries of companies shift, allowing tasks that once had to be completed internally to be done externally.

In the past, artisans operated through multi-node supply chains, with each craftsman taking a share of the value, and skills passed down through generations of apprentices. The Industrial Revolution compressed this distributed model into factories, which concentrated coordination "under one roof," capturing most of the production value.

The internet and mobile devices once again reduced matching and coordination costs, giving rise to the gig economy (Uber, DoorDash) and the creator economy: ordinary people with a camera and an internet connection began to take on tasks that previously only studios, publishers, and agencies could handle.

Bridge Class

Before the infrastructure capable of capturing all value emerges, each of the aforementioned transitions will first produce a "bridge class" that proves the new model works.

Artisans demonstrated that distributed production is feasible, and then factories centralized to capture the value; creators proved that individuals could build audiences and generate income at scale, and then major platforms (YouTube, Instagram, Substack) took most of the economic benefits, becoming the default focal points of the entire system.

The bridge class took on the risks of new technologies and validated that demand was real. Once the infrastructure catches up, a new batch of institutions will massively capture the value.

The gig economy and creator economy are the two most recent bridge classes. They have proven that work can be disassembled, distributed, and compensated outside of traditional employment relationships.

However, they still rely on platforms to package these economic activities: using Stripe for payments, YouTube for content distribution, and Uber for ride matching. Coordination costs have decreased but have not disappeared, as the infrastructure for payments and identity still assumes that both parties in a transaction are human.

Programmable Labor Meets Programmable Money

We are now in the early stages of the next transformation, which depends on two things being in place simultaneously.

The first is programmable labor. AI agents are a new class of labor participants, unrestricted by working hours, headcounts, or geography, scaling through computational power rather than hiring.

A top-level agent can decompose tasks, delegate them to specialized sub-agents, evaluate their outputs, and arrange the next steps, all without human intervention. At this point, the basic unit of labor is no longer positions, working hours, or even deliverables, but the tasks themselves.

In the past, humans packaged tasks into jobs, jobs into careers, and careers into companies simply because that was the only organizational form available at the time. Once you can directly price individual tasks and assign them, "packaging" shifts from a structural necessity to an option.

The second is programmable money. Today, stablecoins represent an asset class of about $300 billion, with credible predictions from multiple institutions suggesting it could surge to $2 trillion in the coming years. Stablecoins compress the entire payment supply chain into a programmable transaction.

The gig economy has not fully disassembled labor because you still rely on Stripe, PayPal, or bank accounts at both ends of the transaction, and these infrastructures presuppose an ongoing relationship between known parties.

Stablecoins may be the best solution prepared for this new class of labor agents. An agent can pay another agent based on output, with amounts as small as a fraction of a cent, and settlements completed within 500 milliseconds, without the need for accounts, invoices, or any intermediaries.

Meta recently began distributing USDC to creators on Polygon and Solana, while AWS launched AgentCore to support stablecoin micropayments specifically for transactions between agents. These are early signals that the world's largest tech companies view stablecoins as the settlement layer for the next generation of economic activity.

The combination of programmable labor and programmable money creates the possibility for the first time in history: a production line without organizational entities, no companies, no payroll systems, no human resources departments, just a series of tasks dispatched, executed, priced, and settled at machine speed.

This is the true disassembly of labor.

Practical Application Scenarios

Merit Systems has developed a product called Poncho that makes all of this very concrete. Poncho provides AI agents with a wallet.

With it, agents can bypass paywalls, access advanced tools, and pay for services, only paying for the actual usage. Poncho integrates payment protocols like x402 and MPP, embedding payment authorization directly into HTTP requests: agents see the price, make the payment, and then gain access.

This represents another way for economic value to flow on the internet. Agents no longer need to subscribe to a large bundle of services that may or may not be useful; instead, they can precisely pay for the specific data, API calls, or computational power needed to complete a particular task.

The early internet explored this idea under the banner of "micropayments," but it never took off. One reason is that credit card fees economically cannot support such small payments, not to mention a host of other challenges, and there was no internet-native payment track at the time.

Stablecoins, leveraging infrastructures like Solana and Ethereum, can complete settlements instantly for just a fraction of a cent, meaning pricing can finally align with the granularity of work.

Repackaging

If you follow this hypothesis further, work will increasingly be paid for by agents to complete tasks for other agents, leading to a change in the form of companies. You no longer need to internalize every function.

What you truly need to excel at is clearly defining what needs to be done, what standards to use to measure quality, and how to ensure these outputs combine to create a whole greater than the sum of its parts.

This also extends to the creator economy. Peer-to-peer tipping has always struggled to gain traction, as evidenced by Clubhouse and Farcaster. But micropayments are particularly suitable for interactions between machines: small payments carry no social awkwardness and no expectation of reciprocity.

If agents become the primary consumers of digital content, the subscription models and paywalls that have long dominated the internet may give way to per-use billing executed by programs.

As AI-generated content floods various channels, the premium on human judgment and craftsmanship will only increase, and the most interesting business models will emerge at the intersection of human taste and machine execution.

In an economy driven by agents, the role of humans is to repackage labor. You are the orchestrator. Your job is to design a system that allows different agents to perform their roles according to specific configurations, creating a flywheel that gradually pushes out the desired results.

Your value lies in knowing what tasks to delegate, how to evaluate them, and how to combine them into something that generates compound returns.

Companies will not disappear, but future companies will increasingly resemble an intelligent layer built on top of a global programmable labor market rather than a container for labor.