Natural: Building an Agent-Native Financial Stack

  natural-building-an-agent-native-financial-stack.html
  Business and Finance on Mon, Jul 20th, 2026 by TechCrunch
      Locale: UNITED STATES

The Friction of Human-Centric Payments

For decades, the gold standard of online payments has been the "checkout flow." This process is designed to prevent fraud and verify identity through a series of checkpoints that are fundamentally incompatible with autonomous software. An AI agent tasked with booking a flight, purchasing an API subscription, or hiring a freelance contractor via a digital marketplace cannot easily navigate a CAPTCHA or respond to a SMS verification code sent to a human's phone.

While some developers have attempted to solve this by hard-coding API keys or granting agents access to existing corporate credit cards, these methods are fraught with security risks. Giving an autonomous agent a wide-open line of credit is an invitation for catastrophic failure—either through logic loops that drain funds or vulnerability to prompt injection attacks that could redirect payments to malicious actors.

Natural's Architecture: Payments for Non-Humans

Natural aims to replace this antiquated model with an "agent-native" financial stack. Rather than trying to fit AI agents into human payment pipes, Natural is constructing a system where the agent is the primary account holder. The core of this proposition likely involves the creation of programmable wallets and dynamic spending limits that can be adjusted in real-time based on the agent's specific mission.

By implementing a layer of programmatic governance, Natural allows a human overseer to set strict parameters—such as a maximum spend of 50 per transaction or a total budget of500 for a specific project—without needing to manually approve every single transaction. This effectively moves the "human-in-the-loop" from the transaction level to the policy level. Instead of clicking "Pay Now," the human defines the "Spending Logic," and the agent executes within those boundaries.

Taking on the Incumbents

For years, Stripe has dominated the developer ecosystem by making it incredibly easy for humans to accept payments. However, Stripe's model is built on the assumption that there is a merchant and a customer, both of whom are legal entities with verified identities (KYC).

Natural is betting that the next wave of commerce will be "Agent-to-Agent" (A2A) or "Agent-to-Merchant" (A2M). In an A2A economy, one AI agent might hire another AI agent to perform a micro-task—such as data cleaning or image generation—and pay it instantly in a fraction of a cent. This level of granularity and speed is impossible within the current legacy banking system, which is bogged down by settlement delays and high flat-fee structures.

The Implications of the Agentic Economy

The $30 million infusion of capital signals a broader shift in the AI industry. We are moving away from the "Chatbot Era," where the primary output is text, and entering the "Agentic Era," where the primary output is action. For action to be scalable, it requires a medium of exchange that is as fast and flexible as the AI itself.

If Natural succeeds in decoupling payments from human identity and attaching them to programmatic identity, it will unlock a massive new sector of economic activity. Businesses could deploy fleets of agents to optimize their supply chains, negotiate contracts, and procure services in real-time without a single manual invoice being generated. The transition from "software as a service" (SaaS) to "agents as a service" will depend entirely on the existence of a seamless, secure payment rail that can handle millions of micro-transactions per second without human intervention.