Most people assume the U.S. dollar has always been the world's reserve currency. It hasn't.

Before the dollar took that position in 1944, global reserve currencies rotated across centuries: the British pound, the Dutch guilder, the Spanish real. Each shift tracked the era's dominant economic and technological power. The dollar itself was created in 1792. It took 152 years, two world wars, and the near-collapse of Europe to finally claim the throne.

We treat the current system as permanent. History treats it as a phase.

The Question Nobody Is Asking About AI

AI is no longer just a tool. It writes, negotiates, manages workflows, executes tasks, and increasingly operates without a human in the loop. Autonomous agents are being deployed to handle customer service, run supply chain decisions, and manage financial operations at scale.

Here is the question that almost nobody is asking: when an AI agent completes work, how does it get paid?

Not in theory. In practice. Right now.

A wire transfer requires a bank account, identity verification, and business days. PayPal requires a human-linked account and jurisdiction-dependent rules. A corporate checking account requires legal structure, compliance filings, and a human signatory. None of these were designed for software that operates 24 hours a day, crosses borders instantly, and has no legal identity.

The fiat system was built for humans. It has friction by design. Banks are open certain hours. International transfers take days. Settlement requires intermediaries. That friction was acceptable when the actors in the system were people.

Machines do not wait.

What Bitcoin Was Actually Designed For

Strip away the price speculation and the cultural noise, and Bitcoin was designed to solve exactly this problem.

In October 2008, Satoshi Nakamoto published the Bitcoin whitepaper with a single-sentence framing that has aged better than almost anything written in finance that decade:

"A purely peer-to-peer version of electronic cash would allow online payments to be sent directly from one party to another without going through a financial institution." -- Satoshi Nakamoto, Bitcoin: A Peer-to-Peer Electronic Cash System (2008)

A few months later, in a P2P Foundation post announcing Bitcoin's launch, Nakamoto added:

"The root problem with conventional currency is all the trust that's required to make it work." -- Satoshi Nakamoto, P2P Foundation (February 2009)

Neither of those sentences mentions humans. They describe a system for transacting parties. Any transacting parties.

Bitcoin does not require identity. It asks one question: do you have the key? Any entity that holds a private key can send and receive value. A software process qualifies. That is not a metaphor or an aspirational future state. It is how the protocol works today.

This Is Already Happening

In November 2024, Lightning Labs released an open-source toolkit that allows AI agents to operate natively on the Bitcoin Lightning Network. No bank account. No API keys tied to a human identity. No registration. The agent runs a Lightning node, pays for services, and hosts paid endpoints autonomously.

Elizabeth Stark, CEO of Lightning Labs, described the intent directly: "We're creating infrastructure for the future of autonomous commerce. This toolkit enables AI systems to participate directly in economic activities using Bitcoin's secure, decentralized network." -- Elizabeth Stark, Lightning Labs (November 2024)

This is not a proof of concept. Coinbase has launched Agentic Wallets. Stripe has previewed machine payments for USDC. CoinGecko activated Lightning-based API endpoints priced per request. The pattern is consistent: the payment infrastructure being built for autonomous software is crypto-native, not fiat-native.

Here is what that looks like in practice: an AI agent needs data to complete a task. It calls a paid API. The API charges 0.01 cents per request. The agent pays via Lightning in milliseconds, receives the data, and continues. No checkout flow. No human approval. No billing cycle. The entire loop runs inside a single execution.

Traditional payment rails cannot do this. Stripe's minimum transaction fee makes sub-cent payments economically unviable. Wire transfers would take longer than the task itself. The friction that humans tolerate because we have no alternative is a hard technical blocker for autonomous systems running at machine speed.

The Reserve Currency Argument

Here is the actual claim: the next reserve currency may not be chosen by governments or central banks. It may be selected by autonomous systems optimizing for transaction efficiency.

The dollar became the global reserve because it offered stability when the world needed it most. Post-World War II, European currencies were devastated, gold was impractical to move at scale, and the U.S. had the productive capacity and institutional strength to back a global standard. The Bretton Woods system formalized what economic reality had already decided.

What economic reality is now deciding is different. The actors accumulating and transacting value at the highest velocity are not nation-states. They are software systems. And those systems have no loyalty to the dollar, no bias toward fiat, and no preference for any currency that does not minimize friction and maximize programmability.

Most people assume the next reserve currency will emerge from geopolitical competition -- a digital yuan, a BRICS alternative, or a reformed IMF instrument. What actually happens is that the selection mechanism has already changed. Humans trusted institutions. Machines verify cryptographic proof.

The Tradeoff

None of this means Bitcoin wins outright. The tradeoff is real.

Bitcoin's volatility makes it a poor unit of account for agents pricing services in dollars. That is why stablecoins on Lightning are growing alongside Bitcoin itself -- agents can transact in dollar-denominated value while settling on a permissionless rail. The architecture may end up being: Bitcoin as the base layer, stablecoins as the transaction medium, Lightning as the settlement network.

What is not in question is whether autonomous software needs a payment system that fiat cannot provide. That gap exists today. The infrastructure being built to fill it is not a bank. It is a blockchain.

If you are planning the next decade on the assumption that financial systems will remain human-centered, you are not just late. You are solving for the wrong actor.

About the Author

Martin Goetzinger has spent his career in enterprise software sales, helping large organizations such as Apple, Microsoft, and Verizon connect data, insight, and action. His work focuses on transforming how businesses measure success and create customer value through technology.

Outside the enterprise world, he writes about the five forces he believes are reshaping everything: AI, blockchain, energy, personalized health, and robotics. Not from a purely technical lens, but from a human one as to how these technologies will redefine work, wealth, and well-being.

He is based in the U.S. and publishes at www.MartinGoetzinger.com.

Disclaimer

The views expressed in this article are the personal opinions of the author and are provided for informational and educational purposes only. Nothing in this article constitutes investment advice, financial advice, legal advice, or any other form of professional advice. Do not make investment or financial decisions based on the content of this article. Always consult a qualified professional before making decisions that affect your finances, business, or livelihood.