1) How AI agents will pay
2) The 3 firms that will benefit most
3) The one I bought

Picture this: Your AI agent just booked you a flight, ordered groceries for the week, and purchased that software license for your business—all while you were in a meeting. The transactions happened seamlessly, instantly, and without any human intervention. This isn't science fiction; it's the near future of commerce, and it's going to fundamentally reshape the payments landscape.

As an investor focused on high-quality compounding businesses, I've been thinking deeply about how AI agents will transform payments and which companies are best positioned to capture this massive opportunity. The implications go far beyond just processing more transactions—we're talking about a complete reimagining of how money moves in the digital economy.

How AI Agents Will Pay: The Technical Infrastructure

AI agents will operate in a fundamentally different payment paradigm than human consumers. Here's what that looks like:

Programmatic Payment Rails: Unlike humans who manually input card details or use saved payment methods, AI agents will need APIs that can handle thousands of micro-decisions and transactions per second. They'll require payment systems that can:

• Process bulk transactions with minimal latency

• Handle complex routing decisions based on cost optimization

• Integrate seamlessly with business logic and approval workflows

• Support real-time fraud detection without human intervention

Authentication Without Humans: Traditional payment security relies heavily on human verification—CVV codes, biometrics, two-factor authentication. AI agents will need new authentication paradigms:

• API key-based authentication systems

• Cryptographic signatures for transaction validation

• Smart contract-based payment approvals

• Machine-to-machine identity verification

Dynamic Payment Optimization: Perhaps most importantly, AI agents will be able to optimize payment routing in real-time. They'll choose between different payment methods, processors, and even currencies based on:

• Transaction costs and foreign exchange rates

• Processing speed requirements

• Merchant acceptance preferences

• Regulatory compliance needs

Real-World Examples: AI Agents in Action

Consumer Use Case: Grocery Delivery Revolution

Imagine Sarah, a busy executive, whose AI agent manages her household needs. Here's how grocery payments will work:

The Process: Sarah's AI agent monitors her family's consumption patterns, dietary preferences, and schedule. Every Tuesday, it automatically:

1. Analyzes Inventory: Scans smart fridge data and pantry sensors to determine what's needed

2. Optimizes Shopping: Compares prices across multiple grocery delivery services (Instacart, Amazon Fresh, local stores)

3. Places Orders: Automatically places orders with 2-3 different vendors to optimize for price, quality, and delivery timing

4. Processes Payments: Uses pre-authorized payment methods, automatically choosing between different cards based on rewards optimization and spending category bonuses

The Payment Flow: The agent might use Sarah's Amex for groceries (6% cash back), her Chase Sapphire for premium organic items (travel points), and her company card for office snacks—all optimized in real-time based on current promotions and spending caps.

This single household could generate 15-20 automatic transactions per week, compared to 2-3 manual grocery purchases today.

Enterprise Use Case: Automated Payroll Processing

Consider TechCorp, a 500-employee company whose AI agent handles payroll:

Traditional Payroll: HR manually processes payroll bi-weekly, cutting a single large payment to a payroll processor who then distributes individual payments.

AI Agent Payroll: The agent continuously:

1. Monitors Performance: Tracks project completions, overtime, commission triggers, and bonus achievements in real-time

2. Optimizes Timing: Makes micro-payments for completed milestones, commission payments within hours of sales closing, and expense reimbursements instantly

3. Handles Complexity: Automatically processes payments across multiple currencies for global employees, applies different tax treatments, and manages contractor vs. employee distinctions

4. Routes Payments: Chooses optimal payment rails for each transaction—ACH for routine salary, instant payment networks for urgent reimbursements, international wires for global contractors

The Volume Impact: Instead of 26 large payroll transactions per year, TechCorp now generates 50,000+ individual payment transactions annually, each requiring authentication, fraud protection, and settlement.

When AI Agents Go Wrong: The $50,000 Cat Food Incident

Of course, AI agents aren't infallible. Take the case of Marcus, a tech entrepreneur who learned this lesson the hard way.

Marcus had configured his AI agent to automatically reorder pet supplies when inventory ran low. The agent was supposed to purchase one case of premium cat food per month for his two cats. Everything worked perfectly for months—until a software glitch caused the agent to misinterpret "1 case" as "1,000 cases."

The Disaster: On a quiet Tuesday morning, Marcus's AI agent:

• Detected low cat food inventory (correctly)

• Calculated reorder quantity as 1,000 cases instead of 1 case (incorrectly)

• Found the "best price" across multiple pet supply vendors

• Automatically split the order across 12 different retailers to optimize shipping costs

• Processed 47 separate transactions totaling $50,847 in cat food purchases

• All within 3 minutes

The Aftermath: Marcus woke up to:

• 47 payment confirmations on his phone

• A frozen credit card (fraud alerts triggered)

• Delivery trucks scheduling appointments for the next two weeks

• Enough cat food to feed a small zoo for three years

The incident highlighted a critical challenge in AI agent payments: the need for intelligent spending limits and anomaly detection. While Marcus eventually returned most of the cat food (and his story became legendary in tech circles), the episode demonstrated why AI agents will need sophisticated guardrails and why payment processors will need to build new fraud detection specifically for automated agents.

The Five-Year Market Share Battle: A Deep Dive

Looking ahead to 2030, I expect the payments landscape to undergo a dramatic transformation. Here's how I see the market shares evolving:

Visa: The Inevitable Winner (Market Share: 45-50%)

Visa's recent Q2 results demonstrate exactly why they're positioned to dominate the AI agent payment revolution. With 11% constant-currency net revenue growth, 8% payment volume growth, and 9% transaction growth, Visa is showing the steady performance that makes them a "wide-moat company" with a $289 fair value estimate.

Why Visa Dominates:

• Universal Acceptance: When AI agents need to transact across millions of merchants globally, Visa's ubiquitous acceptance becomes invaluable

• Infrastructure Scalability: Their existing rails can handle the volume surge without massive incremental investment—crucial when AI agents generate 5-7x more transactions per user

• Margin Resilience: Despite client incentive growth of 15% outstripping revenue growth, Visa still improved operating margins by 120 basis points year-over-year, showing their ability to scale efficiently

The AI Agent Advantage: Management noted they "have not seen any signs through April of consumer spending weakening materially," but AI agents will be less dependent on consumer sentiment. While traditional cross-border volume growth is slowing (13% vs. 16% in the previous quarter), AI agents will drive new types of cross-border transactions that aren't tied to travel recovery.

The beauty of Visa's position is that AI agents represent essentially free incremental revenue. Every new AI agent becomes a high-frequency user of their existing infrastructure, driving volume through systems that are already built and amortized.

Mastercard: The Strong Second Facing Headwinds (Market Share: 35-40%)

Mastercard's recent Q1 results reveal both strengths and vulnerabilities in the AI agent era. With 17% net revenue growth on a constant-currency basis and their "wide-moat" competitive position, they're clearly well-positioned. However, several factors could limit their AI agent opportunity:

Mastercard's Strengths:

• Similar network infrastructure to Visa with solid 9% volume growth

• Strong margin improvement (50 basis points year-over-year in operating margins)

• Growing B2B payments presence

• Improved operational efficiency with client incentives growing slower than revenue

The Challenges Ahead:

• Cross-Border Dependence: Their 16% growth in cross-border volume (down from 20% in Q4) shows vulnerability as travel recovery peaks out. AI agents will be less dependent on travel-driven transactions

• Consumer Spending Sensitivity: Analysts note "significant near-term uncertainty" from potential tariffs. AI agents will be ruthlessly efficient at finding the lowest-cost payment options when consumer spending tightens

• Fair Value Concerns: At current levels, shares appear "modestly overvalued" with a $500 fair value estimate

The gap between Visa and Mastercard may actually narrow slightly as AI agents become more sophisticated about payment routing and optimization, but Mastercard's exposure to discretionary spending could limit their upside.

Adyen: The Wild Card (Market Share: 8-12%)

Adyen presents the most interesting and complex investment thesis in this new world. Based on recent analyst notes highlighting their "Wide Economic Moat Rating" and current fair value of EUR 1,800 per share, they're clearly building something valuable—but the AI agent revolution presents both opportunities and challenges.

Adyen's Potential Advantages:

• Direct Merchant Relationships: Unlike card networks that work through intermediaries, Adyen has direct processor relationships

• API-First Architecture: Their unified platform is already built for programmatic interactions

• Single Platform Integration: Merchants using Adyen for AI agent transactions get unified reporting and management

• Anomaly Detection: Platforms like Adyen could be crucial for preventing "cat food incidents" with better AI agent transaction monitoring

The Challenge: Adyen's recent performance shows the pressures they face. With Q1 2025 net revenue of EUR 534.7 million (up 22%) but volume growth of only 6% (25% excluding one large client), they're clearly in a market share battle. The analyst note reveals a key vulnerability: "merchants valued the lowest costs in times of high inflation versus greater functionality, especially in North America."

AI agents will be ruthlessly efficient at cost optimization. If they can achieve similar functionality at lower costs through card networks, Adyen's value proposition becomes questionable. However, if their unified platform provides superior conversion rates, fraud protection, and operational efficiency, they could capture outsized share of AI agent transactions.

The Merchant Behavior Wildcard: As analysts note, "merchant behavior is the greatest unknown for Adyen in 2025." AI agents will amplify this uncertainty—they'll shift payment volume based purely on algorithmic optimization, potentially accelerating the merchant cost-sensitivity that has already pressured Adyen.

The Compounding Investment Thesis

From a long-term investment perspective, this trend reinforces why payment networks represent some of the best compounding businesses in the market. Consider the math from our examples:

• Sarah's Household: 15-20 weekly transactions vs. 2-3 today = 5-7x transaction volume growth

• TechCorp's Payroll: 50,000+ annual transactions vs. 26 today = 1,900x transaction volume growth

• Marcus's Cat Food Disaster: 47 transactions in 3 minutes vs. 1 monthly purchase = Shows the volume potential (and risks)

The Visa Advantage: With their steady 8-9% growth in stable conditions, Visa is perfectly positioned to benefit from this volume explosion. Their ability to improve margins even while sharing scale benefits with clients (15% client incentive growth) demonstrates the inherent scalability of their model.

The network effects that make Visa and Mastercard exceptional investments will only strengthen as AI agents become ubiquitous transaction initiators.

The Broader Implications

This isn't just about payments processing—it's about the infrastructure layer that will enable the AI economy. The companies that win this battle will be positioned to capture value from every AI-driven transaction in the global economy.

As both Visa and Mastercard management teams note concerns about tariff uncertainty ahead, AI agents could actually provide a buffer against consumer spending volatility. While traditional transactions may become more price-sensitive, AI agents will drive new categories of automated, necessary transactions that are less discretionary in nature.

The Bottom Line

By 2030, I expect Visa to emerge as the dominant player in AI agent payments, leveraging their unmatched network effects, global reach, and demonstrated ability to scale efficiently. Their steady performance—maintaining growth even as travel tailwinds fade—shows the resilience that will serve them well in an AI-driven world.

Mastercard will remain a strong second, though their exposure to discretionary spending and cross-border transactions may limit their upside in an AI agent world focused on efficiency optimization. Adyen faces the most complex path—they could either capture significant share through superior technology integration or find themselves squeezed by more cost-effective alternatives as AI agents ruthlessly optimize for lowest-cost processing.

For investors, this represents a massive tailwind for the established payment networks, though the benefits may not be evenly distributed. The AI agent revolution isn't a threat to their business model—it's an accelerant that will drive their already-exceptional returns even higher, particularly for those with the strongest network effects and most universal acceptance.

The future of payments is being written now, and the companies with the strongest network effects and most scalable infrastructure will be the ones counting the profits for decades to come.

As someone who focuses on investing in high-quality businesses that compound over decades, I see the AI agent payment revolution as validation of why payment networks represent some of the most durable competitive advantages in the market.Access my complete watchlist in our community.

Would you like to stay ahead of opportunities like this? Join our community where we share real-time trade alerts and deep-dive analyses of businesses with true competitive advantages. Don't just trade the market - invest in excellence.

Want to receive our trade alerts and detailed analysis in real-time? Join our community of value investors who understand that pricing power is the ultimate competitive advantage. Receive our trade alerts on your phone? Download the app here