The AI-Native Merchant Gold Rush: New Business Models Your ISO Isn’t Built to Serve

What Is an AI-Native Business?

An AI-native business is a company that was founded on and built around artificial intelligence capabilities. It is not a traditional business that has added AI as a feature — it is a business whose core product, delivery model, and customer interaction are fundamentally structured around AI. This distinction matters enormously for payment design, because AI-native businesses have transaction patterns, billing models, and customer relationships that are structurally different from anything the traditional merchant services industry was built to handle.

The clearest indicator of an AI-native business is its billing model. A traditional SaaS company charges a monthly subscription: $50/month for project management software, $200/month for an accounting platform. An AI-native business often charges differently: per usage unit (per API call, per document processed, per image generated), or as a hybrid (a base subscription plus per-usage overage charges). This hybrid billing model creates transaction patterns — potentially thousands of micro-transactions per customer per month — that cannot be processed efficiently using traditional card payment infrastructure.

AI-native businesses are emerging across every vertical. Legal technology companies that charge per brief filed or per deposition transcript processed. Medical billing AI platforms that charge per claim reconciled. AI code generation tools that charge per line of code written. AI creative platforms that charge per image or video generated. AI writing tools that charge per word or per document. Each of these categories is spawning dozens of new companies — and each of these companies needs payment infrastructure that can handle usage-based billing, high-frequency micro-transactions, and AI agent settlement. This is not a niche. It is the next generation of small and medium business.

The volume is building faster than most ISOs realize. McKinsey projects that AI-related SMB formation will generate an estimated $1.5 trillion in new annual payment volume by 2028. Goldman Sachs estimates that AI-native companies currently process approximately $340 billion annually across all payment rails, and that this figure is growing at approximately 47% year-over-year — faster than any other business category. These are not projections for 2035. These are the numbers being reported now. The merchants exist. The volume is real. The question is whether the ISO channel will respond in time.

1. AI-First Software: Why Traditional Merchant Products Cannot Serve AI-Native Businesses

The micro-transaction problem: An AI code generation tool might process 50,000 API calls in a single day for a single enterprise customer. At $0.01 per call, that is 50,000 individual charges worth $500 — a volume and frequency that traditional payment infrastructure was never designed to handle efficiently. Stripe estimates that micro-transactions under $0.50 carry effective processing costs of 10-15% of the transaction value when all fees are included. For AI-native businesses that bill per unit, this makes card processing economically prohibitive. They need an alternative — and most ISOs have not built one.

The subscription-hybrid model problem: AI-native businesses frequently use a hybrid model: a base monthly subscription that covers a baseline of usage, plus per-unit charges for usage above the threshold. This billing architecture requires real-time usage metering, tiered pricing, and dynamic invoice generation — capabilities that standard merchant accounts do not provide. The billing logic lives in the AI company’s software; the payment rail needs to interface with it seamlessly. Most traditional merchant products cannot bridge this gap without significant custom integration work.

The AI agent settlement problem: As AI agents become more capable — and as agentic commerce develops — businesses will need to pay other AI agents for services performed. An AI agent that research a market and deliver a report will need to be paid by the AI agent that commissioned the work. This agent-to-agent payment flow requires a payment infrastructure that does not yet exist in any meaningful way in the ISO channel. The ISO that builds or partners to provide agent settlement rails will be positioned at the foundation of a new transaction category.

2. New Payment Models: Subscription, Usage-Based, and Agent Settlement

Three payment models are emerging as the standard for AI-native businesses, and each requires a different approach from the merchant services industry.

Subscription: A predictable monthly or annual charge for access to an AI platform. This is the most traditional model and the easiest for existing merchant products to handle. However, the competitive dynamics are intense: Stripe and Square have built dominant subscription billing products, and AI companies increasingly expect the same level of sophistication from their payment provider.

Usage-based billing: Charges tied directly to consumption — per API call, per document, per image, per hour of AI compute. This model requires integration between the billing system and the payment rail, real-time metering, and the ability to handle potentially thousands of transactions per customer per day. It is the highest-volume, highest-complexity model — and the one most underserved by the current ISO channel.

Agent settlement: AI agents transacting with other AI agents on behalf of humans. This is the most speculative of the three models — but it is not science fiction. Agentic commerce platforms are already processing billions in annual volume. Stripe, PayPal, and several fintech startups are building agent settlement infrastructure. The ISO channel is not currently part of this conversation — and that is a significant strategic oversight.

3. Emerging Verticals: Where the AI-Native Merchants Are Appearing

AI-native businesses are not a single market — they are a constellation of vertical-specific niches, each with distinct payment requirements, customer acquisition patterns, and growth trajectories. Understanding the vertical landscape is essential for ISOs who want to develop a strategy for this segment.

Legal tech: AI contract analysis, brief generation, deposition transcription, and case research platforms. Billing models: per-document, per-page, or per-case. Volume: high-frequency micro-transactions for high-volume practices. Average transaction value: $0.50–$5.00 per document processed.

Healthcare AI: AI diagnostic assistants, medical billing reconciliation, AI-powered patient scheduling, and clinical documentation tools. Highly regulated environment — compliance requirements (HIPAA, state regulations) create barriers to entry for ISOs who have not built for healthcare. But the payment volume is enormous: US healthcare AI alone is projected to reach $102 billion by 2028.

AI developer tools: Code generation, automated testing, AI code review, and infrastructure automation platforms. The largest and most mature AI-native vertical. Developers pay per API call or per seat. Most transactions are under $1.00 — but volume is extraordinary. Stripe processes over $5 billion annually in AI developer tool payments alone.

AI creative tools: Image generation, video generation, audio synthesis, text-to-speech, and music generation platforms. Per-generation or per-minute billing. High volume, lower individual transaction values. Dominated by consumer-facing platforms but increasingly serving professional creative agencies.