AI and Stablecoins: The Biggest Threat to Global Payment Giants?

Visa stocks down 4.6%.
Mastercard stocks down 5.7%.
American Express stocks down 7.2%.
Capital One stocks down 8.8%.
The market is beginning to price in a structural change. And the concern is very simple.
AI systems do not choose payment methods based on brand or existing infrastructure. They automatically select the fastest and cheapest way to process transactions.
Currently, card payments typically cost merchants between 2% and 3.5% per transaction. Cross-border payments often exceed 4% when including exchange rate differences and intermediary fees.
If AI agents can pay instantly with stablecoins at nearly zero cost, traditional payment systems will lose their expensive advantage.
And payments are at the core of almost every industry. Every business relies on transferring money. That’s why stablecoins are becoming hard to ignore.
Traditional payment systems still face significant barriers.
Card networks charge based on a percentage rate. International transfers can cost hundreds of dollars. Payment delays slow down cash flow between businesses and supply chains.
Stablecoin networks are changing that structure.
Money transfer transactions are settled within seconds or minutes. Cross-border payments cost only a few dollars. Network fees can drop to just a few cents while remaining operational continuously without interruption.
On a global scale, these differences become very significant. According to World Bank data, global remittance fees still average 6.6%.
Now, combine that with the scale of global payments.
Just the B2B payment flow exceeds $1.6 quadrillion annually. Even small efficiency improvements shift trillions of dollars.
Data on adoption reflects this transition.
Stablecoin transaction volume is projected to reach about $33 trillion by 2025, an increase of over 70% from the previous year.
Total supply has expanded to over $300 billion, compared to around $10 billion just a few years ago.
Estimates from Citi suggest the supply could reach $1.9 trillion by 2030 and possibly $4 trillion in an optimistic scenario.
At that scale, stablecoin issuers could become some of the largest buyers of U.S. Treasury bills worldwide.
This also puts pressure on banks.
Banks rely on deposits to fund lending activities. Instead, stablecoins hold reserves directly in the form of Treasury bills.
If companies start holding operational capital as stablecoins instead of bank deposits, part of the funding base supporting traditional lending begins to shift.
Regulators are paying attention.
In recent U.S. cryptocurrency regulation discussions, banking groups have strongly opposed allowing stablecoins to generate profits.
The concern is clear. Banks supported by digital dollars and Treasury bills earning outside profits could accelerate the shift of deposits.
Artificial Intelligence (AI) adds another layer of acceleration.
Payments are increasingly shifting from humans to software systems.
AI agents automatically make payments via APIs.
Real-time resource rental software.
Machines continuously pay for services.
These systems optimize strictly for cost and speed.
When AI compares card fees based on percentages with near-instant stablecoin payments, routing decisions become more machine-driven than behavior-based.
Financial institutions are already preparing for this possibility.
Fireblocks’ research shows nearly half of organizations have used stablecoins for payments, while over 80% report infrastructure readiness.
McKinsey estimates that actual stablecoin payments in payroll, remittances, and B2B transactions reached nearly $390 billion annually and are growing rapidly.
Even Visa and Mastercard are now integrating stablecoin payment infrastructure.
The payment network will not disappear overnight.
But the market may already be pricing in a future where transferring money becomes significantly cheaper.
And that directly challenges one of the most profitable segments in global finance.