Natural raises $30M to build payments for AI agents

Key Takeaways

• Natural raised $30M Series A led by Forerunner to build payment rails designed for autonomous AI agents
• Current payment infrastructure requires human authorization, creating bottlenecks for AI agents executing transactions
• The startup is competing directly with Stripe, which is also racing to redesign payment systems for AI

Natural, a startup building payment infrastructure for AI agents, has raised $30 million in Series A funding led by Forerunner. The round brings total funding to $40 million and sets up a direct confrontation with Stripe, which dominates online payments but was built for humans, not autonomous software.

The core problem: AI agents can now research vendors, compare prices, and negotiate deals. But when it comes time to pay, they need a human to authorize the transaction. Credit cards, ACH transfers, and other payment rails assume someone is clicking "confirm." That assumption breaks when the buyer is software.

Why existing payment rails fail AI agents

Traditional payment systems were designed around human authorization. Every credit card transaction expects a cardholder to enter a CVV or tap a phone. ACH transfers require approval through banking apps. These checkpoints exist for fraud prevention and regulatory compliance, but they create latency that defeats the purpose of autonomous agents.

Consider an AI agent managing logistics for an e-commerce company. It identifies a freight vendor, negotiates a rate, and confirms delivery timing. The entire process takes seconds. But authorizing payment? That waits for a human to wake up, check their email, and click through a banking interface. The bottleneck is not the agent's capability. It's the payment infrastructure.

Natural is rebuilding the system from scratch. The startup describes itself as an "agent orchestration layer" that lets AI agents move and store funds autonomously. Companies integrating Natural's infrastructure can let their agents make payments, collect funds, and transact with other agents without human intervention at every step.

The founding team's fintech background

CEO Kahlil Lalji did not want to start another fintech company. His previous startup Ivella, a YC-backed banking product for couples, sold to Earnin in 2023. After two years as an engineer there, he told TechCrunch he had been "burned by the finance sector" when the zero-interest-rate era ended.

But the opportunity was too obvious to ignore. "I kept on coming back to it," Lalji said. "It just feels obvious that agentic payments are going to be structurally the most important problem in the space."

Lalji recruited Eric Wang, his Ivella co-founder, and Walt Leung, a former engineering manager at Nextdoor. The team founded Natural in 2025.

Competing with Stripe on speed

Stripe is not standing still. The payments giant is also racing to redesign its infrastructure for AI agents. But Natural is betting that its greenfield architecture gives it an edge over a company maintaining legacy systems built for a different era.

"We've made enough critical architectural decisions to give us a good shot at competing with incumbents," Lalji told TechCrunch. The startup has been operating in beta, and its fast development speed is the key competitive advantage he's counting on.

The mission has attracted senior hires from Stripe, Ramp, and Square. Fintech veterans joining a pre-revenue startup signals that insiders see the opportunity as real.

The stablecoin question

Natural is not alone in this space. Skyfire Systems, backed by DCVC, is building agent payment infrastructure using USD-backed stablecoins. The crypto approach sidesteps some legacy banking friction, but it introduces regulatory complexity and limits which merchants can accept payments.

Natural plans to support both: stablecoins and traditional bank payments. That dual-track approach expands the addressable market but increases engineering complexity.

Why Forerunner led the round

Kirsten Green, Forerunner's founder and managing partner, led Natural's Series A. Her firm focuses on consumer experiences and the future of commerce. Green was drawn to Natural's broader ambitions beyond just enabling checkout. The startup is also tackling how disputed transactions are handled when the "customer" is software.

Chargebacks and disputes are a $40 billion annual problem for merchants. When an AI agent makes a purchase, who is liable for fraud? How do refund policies apply? Natural is building answers into its architecture rather than patching them on later.

The market size argument

Lalji's boldest claim: the total number of payments could increase by two to four orders of magnitude if transactions happen at computer speed instead of human speed. That's 100x to 10,000x more transactions.

The math depends on AI agents becoming ubiquitous and trusted to spend money. If that happens, a single company might generate thousands of micro-transactions daily as agents negotiate, procure, and pay for resources in real time. Each transaction would be smaller, but the aggregate volume would dwarf today's payments market.

Whether enterprises will trust autonomous agents with spending authority remains an open question. But the infrastructure needs to exist first. Natural is betting that building the rails creates the conditions for adoption.

Frequently Asked Questions

What is Natural building for AI agents?

Natural is building payment infrastructure that allows AI agents to autonomously move money, collect payments, and transact with other agents without requiring human authorization at each step.

How does Natural compete with Stripe?

Natural is building greenfield infrastructure designed specifically for AI agents, while Stripe must adapt existing systems built for human-initiated transactions. Natural is betting its speed of development will outpace Stripe's ability to retrofit.

Who funded Natural's Series A?

Forerunner led the $30 million Series A round, bringing Natural's total funding to $40 million.

Does Natural use stablecoins for payments?

Natural plans to support both stablecoins and traditional bank payments, unlike competitors like Skyfire Systems that focus primarily on crypto-based rails.

When will AI agents routinely make autonomous payments?

The timeline depends on enterprise trust and regulatory clarity. Natural is building infrastructure now, anticipating that agent payment volumes could increase by 100x to 10,000x once adoption accelerates.

Source: Venture Capital News | TechCrunch / Marina Temkin